Colin J. Bennett
Associate Professor
Department of Political Science
University of Victoria
Victoria, B.C. V8W 3P5
CJB@Uvic.Ca
Paper prepared for the Center for Social and Legal Research, DOE Human Genome Project:
Final Draft, April 1995.
I. PRIVACY, PUBLIC POLICY AND POLITICAL SCIENCE
II. PRIVACY AND THE POLITICAL THEORY OF THE DEMOCRATIC STATE
III. PRIVACY, POLITICAL CULTURE AND POLITICAL IDEOLOGY
IV. PRIVACY AND THE POLITICS OF SURVEILLANCE
V. PRIVACY, DECISION-MAKING AND THE IMPACT OF NEW INFORMATION TECHNOLOGIES
VII. PRIVACY, CAPITALISM AND THE INFORMATION MARKET
VIII.THE POLITICAL ECONOMY OF GENETIC INFORMATION
IX. CONCLUSION: POLITICAL ECONOMY AND THE PRIVACY DEBATE
It is an almost customary feature of any analysis of privacy to begin with a disclaimer about the inherent difficulty, perhaps impossibility, of defining exactly what "privacy" is, and of disaggregating its various dimensions. Many attempts have been made; those efforts still continue (e.g. Powers, 1993). For the purposes of this review, however, it would be misleading and confining even to try to provide a general definition of "privacy" to focus the analysis. All definitions, to some extent, are based on questionable assumptions about individualism, and about the distinction between the realms of civil society and the state. Many gloss over essential cultural, class-related or gender differences. It is those very assumptions that require careful interrogation if the "politics" of privacy are to be unearthed.
Successful attempts have probably been made to specify the various roles that privacy may perform within modern political systems (Westin, 1967: 32-39; Flaherty, 1989: 8). A useful distinction can also be made between privacy as an intrinsic or aesthetic value, or the "restriction of personal information as an end in itself" (Rule et al, 1980: 22), and privacy as an instrumental or "strategic" value where the aim is perhaps to ensure that the "right data are used by the right people for the right purposes" (Sieghart, 1976: 76). There is also, of course, a distinction between the claim to privacy and a "right of privacy" (McCloskey, 1980). Over thirty years of semantic and philosophical analysis, however, leaves this reviewer with the overwhelming sense that privacy is a deeply and essentially contested concept (see, Schoeman, 1984).
So too is the word "politics." Political scientists have struggled for years to place some conceptual boundaries around their discipline. Thus we have been told that politics is the study of "who gets what, when, how" (Lasswell, 1950), or that it is the "authoritative allocation of values" (Easton, 1965), or that it is present whenever there is a will to conciliate, compromise or debate (Crick, 1964). But there is really no consensus about the proper focus of the discipline with the result that it has been said that political scientists tend to sit at "separate tables" (Almond, 1988): just as in a restaurant the participants enjoy a passing recognition of the conversations at other tables, for the most part the discussions are confined and unpenetrated by outsiders.
Thus, any claim to define either politics or privacy raises more questions than it answers. We are dealing with two inherently problematic and contestable concepts.
This review is titled the political economy of privacy to capture the broad range of empirical and theoretical questions that are raised for democracy on the one hand, and capitalism on the other. The issue of personal privacy, as it has been framed within advanced industrial states, raises a number of fascinating political questions that overlap the disciplines of political science and economics. The starting point, therefore, is not an abstract and inevitably contestable conception of privacy, nor an artificial definition of disciplinary boundaries, but rather the theoretical and empirical questions that have been, or should be, raised about privacy for any scholar of the politics and economics of advanced industrial states.
I. PRIVACY, PUBLIC POLICY AND POLITICAL SCIENCE
The initial specification of the relevant questions for the political analyst is difficult. There is no tradition of studying information privacy within the academic disciplines of political science or economics. There is, therefore, no identifiable tradition or cumulative literature upon which to draw, at least within these disciplines. I will attempt to explain this shortcoming in the conclusion. Indeed, most of the relevant questions have been posed by scholars affiliated with other academic departments; several sociologists and lawyers will appear regularly in the following review, as will many non-academics, like journalists and civil liberties advocates. Moreover, the relevant literature is sprinkled throughout a range of academic journals, books, conference papers and more journalistic sources.
A cursory review of the way that privacy arises in the standard political science texts and journals suggests that when the concept does appear, it is used to designate policy issues that tend to be peripheral to the major concerns of the "privacy scholars." The bibliographic source ABC Pol Sci lists on average 10 articles per year on "privacy" in its index. A more careful analysis of these works reveals, however, that the majority are in law journals, and are related to a diversity of issues: abortion, homosexuality, the private lives of public figures, private diplomacy, the "private" roles of women, privatization, employment law and so on. There are few articles which treat privacy as a distinct area of public policy, the conceptualization, formation and implementation of which require political or economic analysis.
Furthermore, very little has been written about "informational" (as opposed to "behavioral") privacy by political scientists. The former, defined originally by Westin (1967: 7) as the "claim of individuals, groups or institutions to determine for themselves when, how and to what extent information about them is communicated to others", serves generally to restrict the analysis to the policy problem that has been created by the emergence of modern information and communication technologies.
This review focuses chiefly on information privacy and will not generally be concerned with the privacy literature associated with the application of the right of privacy to intimate personal decisions about abortion or contraception (e.g. Inness, 1992), nor to forms of sexual behavior (e.g. Samar, 1991), nor to the protection of the private lives of public figures (Warren & Brandeis, 1890). Moreover, many "behavioral" privacy issues (e.g. videosurveillance, polygraph testing, intelligent vehicle highway systems as well as genetic privacy) tend to become defined as information privacy questions. In all circumstances, personal information is collected or inferred as a result of the observation of personal activity or behavior. The subsequent storage, use and communication of that information raises a perennial set of information privacy claims.
Hence, our understanding of the privacy issue has only been marginally influenced by the questions that most political scientists or economists would find interesting, nor by the approaches that are typically used in these disciplines for the analysis of other pressing policy problems. Moreover, research findings about privacy have not been seen as furthering theoretical developments within these disciplines. Thus, this review has to be organized not around identifiable traditions within a political science or economics literature, but according to some central theoretical questions that political and economic theorizing can and should address. The body of the review is organized around six separate, but interrelated, questions.
First, what does the tradition of political theory contribute to the debate about the appropriate "balance" between privacy interests on the one hand, and wider societal and community obligations on the other? This raises central theoretical questions about the relationship between the individual and the state, about the boundaries of the public and the private, about the essential contest between liberty and equality.
Second, what is the relationship between attitudes toward privacy and political culture and ideology. Here the paper raises questions typically within the realm of comparative politics. Is concern for privacy rooted in more deep-seated cultural orientations towards the state, and is there any evidence of systematic variation across democratic political systems? As a political issue, is it one of the left, the right or neither? What are the ideological underpinnings of the issue?
Third, what can political science tell us about the nature and extent of surveillance? The focus in this section is upon the qualitative and quantitative changes in the power of public and private institutions that new technology practices bring in their wake. What do we know about the nature and source of these trends? What can we say about their impact on individuals and society?
Fourth, what can political science (and particularly the study of bureaucratic behaviour) tell us about the relationship between technology and politics. The attempt to promote greater personal privacy in an age of rapid technological development raises profound questions of interest to students of public administration and organizational theory about the extent to which similar technologies have a deterministic or convergent impact on organizational choice. The issue raises in a different form the central epistemological question about the relation between structure and agency in politics.
Fifth, privacy can also be seen as a regulatory issue, the response to which may tell us something about the way different states manage technological change. The issue raises a range of fairly traditional questions about agenda-setting, pressure group politics, the legislative response, the choice of policy instruments and policy implementation, adjudication and evaluation.
Sixth, privacy can be analyzed on the basis of economics or public choice assumptions. In contrast to the view that privacy should be seen as a democratic right that the state has a duty to protect through regulatory policy, others see the privacy of personal information as an inherently individual good that should only be protected through individual demands and organizational responses within the context of a market society.
The next section draws some conclusions from this analysis about how these approaches might contribute to the resolution of privacy tensions in the field of genetic information. To what extent is genetic information new and different, and therefore requiring of fresh analytical approaches and policy solutions?
The paper concludes with some reflections on the nature of the debate. It tries to explain why there are so many silences within the political science literature on these issues, and why we have had to look to others (lawyers, sociologists, journalists etc.) for the analysis of questions that are clearly within the broad realm of political economy.
II. PRIVACY AND THE POLITICAL THEORY OF THE DEMOCRATIC STATE
Whatever the psychological or sociological justifications for the importance of privacy, the contemporary political justifications overwhelmingly rest on general assumptions about the continued viability of a liberal political philosophy and epistemology. It is first necessary to establish the tradition from which the contemporary theory of information privacy was derived, and present the various critiques of this position. I will then discuss how those assumptions have been reflected in the privacy literature.
On an epistemological level, the claim to privacy can only be made from a recognition of a society of relatively autonomous individuals. This has its roots in the 17th century political philosophy of John Locke. It rests on notions of difference; my privacy claims and interests are different from yours. It rests on an atomistic conception of society; the community is no more than the sum total of the individuals that make it up. It rests on notions of a distinction between subject and object. I (with my liberty, autonomy, rationality and privacy) am assumed to know my interests, and should be allowed a sphere of being untouched by others.
In John Stuart Mill's words, there should be certain "self-regarding" activities of private concern, contrasted with "other-regarding" activities susceptible to community interest and regulation. The modern claim to privacy, then, rests on a notion of a boundary between the individual and other individuals, and between the individual and the state. It rests on notions of a distinction between the public and the private. It rests on the pervasive assumption that there is a civil society comprised of relatively autonomous individuals who need a modicum of privacy to make rational self-regarding decisions.
Edward Shils is a twentieth century proponent of this view of privacy. He wrote in the 1950s that privacy is essential for the strength of American pluralistic democracy. It bolsters the boundaries between competing, and countervailing centers of power. It reinforces the barriers between the individual and the state and within the contours of civil society (Shils, 1956: 154-60). Alan Westin (1967: 24) provided an eloquent statement of the importance of privacy for liberal democratic societies. In contrast to totalitarian regimes, "a balance that ensures strong citadels of individual and group privacy and limits both disclosure and surveillance is a prerequisite for liberal democratic societies. The democratic society relies on publicity as a control over government, and on privacy as a shield for group and individual life...Liberal democratic theory assumes that a good life for the individual must have substantial areas of interest apart from political participation."
Westin went on to address the specific functions that privacy plays. It promotes the freedom of association. It shields scholarship and science from unnecessary interference by government. It permits the use of a secret ballot and protects the voting process by forbidding government surveillance of a citizen's past voting record. It restrains improper police conduct such as "physical brutality, compulsory self-incrimination and unreasonable searches and seizures." It serves also to shield those institutions, such as the press, that operate to keep government accountable (1967: 25).
The critique of liberal political theory as a basis for privacy has come from at least five different directions. Some, for example, have noted that there is a definite negative dimension to the notion of privacy as the "right to be let alone." On the one hand, it draws attention to why one might want to be left alone, and to the criticism that privacy rights are predominantly asserted for those that have the most to hide. Here is a quote from an early article by H. W. Arndt (1949: 69): "The cult of privacy seems specifically designed as a defence mechanism for the protection of anti-social behaviour." He goes on to equate privacy with the almost pathological obsession with possessive individualism: "The cult of privacy rests on an individualist conception of society, not merely in the innocent and beneficial sense of a society in which the welfare of individuals is conceived as the end of all social organisation, but in the more specific sense of 'each for himself and the devil take the hindmost'" (1949: 70).
A similar critique of the theory of information privacy was presented in a famous article by Richard Posner (1978b). Posner's central point was that the application of the principle of information privacy has an unfortunate corollary, namely that it allows people to conceal personal information in order to mislead and misrepresent their character. Others, including government institutions, "have a legitimate interest in unmasking the misrepresentation" (1978a: 20). "It is no answer," Posner continues, "that, in Brandeis's phrase, people have 'the right to be let alone.' Few people want to be let alone. They want to manipulate the world around them by selective disclosure of facts about themselves. Why should others be asked to take their self-serving claims at face value and prevented from obtaining the information necessary to verify or disprove these claims?"
A second line of attack has come from those who find the distinction between public and private problematic. Joseph Bensman and Robert Lilienfeld, for instance, contend that the "private and the public are inextricably intertwined and interlaced. They cannot be treated as separate entities" (1979: 182). They are at the very least "complex-structured concepts" that operate on different dimensions. Stanley Benn and Gerald Gaus (1983: 7-11) contend that the public/private distinction must be qualified according to whether one is analyzing the "allocation of access to information, resources etc., the capacities in which agents enjoyed that access, and in whose interest it was used."
Thirdly, the public/private dichotomy (and by extension much theorization about privacy) has also been regarded as inherently gendered. Feminists have critiqued liberalism for reifying a distinction between a private, domestic (female) world, and a public sphere that is chiefly the preserve of men (Pateman, 1983; Allen, 1985). Anita Allen and Erin Mack critique Warren and Brandeis on these grounds: they "were not critical of the ways in which homelife, assertions of masculine personality, and norms of female modesty contributed to women's lacking autonomous decisionmaking and meaningful forms of individual privacy." They advocated "too much of the wrong kinds of privacy -- too much modesty, seclusion, reserve and compelled intimacy -- and too little individual modes of personal privacy and autonomous private choice" (1990: 477).
Fourthly, and from the perspective of democratic theory, some would also contend that the liberalism of Locke and Mill, upon which the theory of information privacy rests, represents just one version of democratic theory. Yet some have argued, such as Carole Pateman (1970), that there are two general traditions of democratic theory: the one is a liberal tradition rooted in 18th century natural rights theory. The other is derived from the view that the test of a democracy is not the protection of individual or minority rights, nor the degree of competition between centers of power. Rather the test is the degree of participation, cooperation, community consciousness and so on, values that are not necessarily promoted by asserting the "right to be let alone." This argument finds current reflection, of course, in the renewed interest in the communitarian theorizing of Amatai Etzioni and which has resonated with contemporary political elites of the left and the right in both Europe and America (Etzioni, 1994).
It may even be argued that some of the most creative civilizations in history (such as in Greece, Rome, Renaissance Italy) flourished despite or maybe because of the lack of individual privacy (Bennett, 1992: 32). Information privacy is, therefore, a precondition not of democracy per se but of a particular form of liberal democracy, the theoretical justifications for which were provided by Locke, Madison, and Mill, rather than by Jean-Jacques Rousseau.
The conceptualization of distinct private and public realms almost inevitably leads the debate to a discussion of how privacy conflicts with social or community values. It leads often to the view that privacy and social values like internal security, social welfare, government efficiency and so on are necessarily antithetical. The problem here is not only the deeply contested and ambiguous quality of these concepts, but also that the promotion of privacy can itself be socially important. This is Priscilla Regan's point:
Most privacy scholars emphasize that the individual is better off if privacy exists. I am arguing that society is better off when privacy exists. I argue that society is better off because privacy serves common, public and collective purposes. If you could subtract the importance of privacy to one individual in one particular context, privacy would still be important because it serves other important functions beyond those to the particular individual (1993a: 16).
A final, and more recent, critique emerges from those who argue from post-structuralist assumptions that the essential ontological premise about the central autonomy of the subject is misguided. In explicating Foucault's (1979) notion of the "Panopticon", as a new form of "everyday surveillance" and social control, Mark Poster (1990: 97-98) explains the post-modern/post-structuralist argument as follows:
Foucault taught us to read a new form of power by deciphering discourse/practice formations instead of intentions of a subject or instrumental actions. Such a discourse analysis when applied to the mode of information yields the uncomfortable discovery that the population participates in its own self-constitution as subjects of the normalizing gaze of the Superpanopticon. We see databases not as an invasion of privacy, as a threat to a centered individual, but as the multiplication of the individual, the constitution of an additional self, one that may be acted upon to the detriment of the "real" self without the "real" self ever being aware of what is happening.
Poster's analysis places the "mode of information" and especially the surveillance capacity of modern information technology at the heart of contemporary social transformations. For him, the theory and language of information privacy is irrelevant. The more profound question for the post-modern era is nothing less than "where the human self is located if fragments of personal data constantly circulate within computer systems, beyond any agent's personal control" (Lyon, 1994: 18).
Thus, the privacy debate has sometimes raised some insightful and controversial theorization about the concepts "public" and "private" and has echoed some of the claims and counterclaims within political theory generally. For the most part, however, the political theorization about privacy has operated within the basic liberal paradigm. The privacy literature, though vast, has remained relatively untroubled by deeper ontological questions about the nature of the "self" in modern, or postmodern conditions. It has assumed a distinction between the realms of the public business of the state, and the private spheres of individual life. It has remained relatively unaffected by deeper questions about cultural relativity, or bias according to class or gender.
To some extent, this is explained by the fact that much of the more philosophical debate about privacy has been directed toward the political and legal arena. It has been said (normally by those trained in European schools) that most American political theory is but a footnote to the Constitution. In the privacy area, however, there is some truth in this. The bulk of the more abstract and conceptual literature was prompted by the need to understand the emerging "right to privacy" that the Supreme Court was in the process of developing and applying to private decisions about intimate family concerns such as contraception and abortion. Much of the philosophy of privacy is, therefore, understandably directed towards emerging legal doctrine (e.g. Prosser, 1960; Fried, 1968; Pennock & Chapman, 1971; Parker, 1974; Gavison, 1980; Parent, 1983).
The pervasiveness of liberal assumptions within the literature has had a number of political and policy implications. First, it explains the continuous reference in the privacy literature to the concept of balance. However conceptualized, privacy is not an absolute right; it must be balanced against correlative rights and obligations to the community. Richard Hixson (1987) conceptualizes "balance" as the "continuing struggle over the meaning of private and public, the jurisprudential debate over individual autonomy and collective welfare, between the person and the state, the individual and the community" (Hixson, 1987: xv-xvi). An assumption of "balance" underlies many of the official investigations into privacy policy. The US Privacy Protection Study Commission, for instance, began its analysis by declaring that "the Commission has constantly sought to examine the balance between the legitimate, sometimes competing, interests of the individual, the record-keeping organization, and society in general" (PPSC, 1977: xv).
The assumption underlies the doctrine of "fair information principles" (FIPs) for the appropriate collection, retention, use and disclosure of personal information. The codification of these principles has varied over time and space. They appear either explicitly or implicitly within all national data protection laws (including those in the US and Canada that are called "Privacy" Acts). They also form the basis of international agreements, such as the 1981 Guidelines from the Organization for Economic Cooperation and Development (OECD, 1981), the 1981 Convention from the Council of Europe (CoE, 1981), and the more recent Directive on Data Protection from the European Union (Council of the EU, 1995; see also Regan, 1994).
I have argued elsewhere (Bennett, 1988a; 1992: 101-111) that, while the codification may vary, the principles tend to boil down to six essential tenets: 1) a principle of openness -- the very existence of personal record-keeping systems should be publicly known; 2) a principle of individual access and correction; 3) a principle of collection limitation -- organizations should only collect and store information that is relevant to accomplish a legitimate task; 4) a principle of use limitation -- data should only be used for purposes consistent with those tasks; 5) a principle of disclosure limitation -- personal data should only be disclosed externally for legitimate reasons or with the consent of the individual; and 6) a principle of security, against loss, unauthorised access, destruction, use, erasure or modification of personal data.
The development of these statutory principles rests on some basic liberal assumptions about procedural justice. We all have privacy rights and interests, it is assumed, but those concerns can only be subjectively defined. No paternalistic definitions of the privacy value appear; no attempts are made to second-guess the privacy interests of individual citizens. A rational faith in the capacity of long-standing principles of procedural justice to counter the worst effects of new technologies has been pervasive.
Further work needs to be done on the political theory of privacy. As a common problem which has arrived on the agendas of different democratic states at roughly the same time, it provides a fascinating opportunity to interrogate wider theories about politics in different advanced industrial states. Unlike some other regulatory issues, it raises central and enduring questions about the power of the state, the respect for civil liberties, the relationship between the state and civil society, the definition of the "subject" within conditions of modernity (or post-modernity).
I would contend, however, that the political theory of privacy (in both the United States and Europe) has largely operated within a liberal paradigm and has not yet confronted more profound ontological and epistemological issues. The links between the vast tradition of political theory (including its rich and multifaceted critique of the many varieties of liberalism) and the theoretical literature on privacy are, therefore, tenuous.
III. PRIVACY, POLITICAL CULTURE AND POLITICAL IDEOLOGY
Although examples of systematic record-keeping by governments can be found in ancient and medieval times, there is common consensus that its rise is most closely associated with the development and bureaucratization of the modern nation-state (Tilly, 1975; Dandeker, 1990). The expansion and institutionalization of state power since the 16th century brought with it the need for more complex, discriminating and formal record-keeping systems, many of which contained personal information. Three types of personal record can be identified: administrative records (generated by a transaction with an agency, such as applying for a grant, reporting your income, applying for welfare, getting married etc.); intelligence records (serving an investigate purpose, such as police files); and statistical records (created through census or survey research methods) (US HEW, 1973). The institutionalization of personal record-keeping is a trend common throughout advanced industrial states.
Westin argued in Privacy and Freedom, however, that different historical and political traditions among Western nations were likely to create different results in the overall balance between privacy and government. Britain exhibits a "deferential democratic balance", a combination in which there is "greater personal reserve between Englishmen, high personal privacy in home and private associations, and a faith in government that bestows major areas of privacy for government operations". West Germany exhibits an "authoritarian democratic balance" in which "respect for the privacy of person, home, office and press still gives way to the claims of official surveillance and disclosure." The United States exhibits an "egalitarian democratic balance, in which the privacy-supporting values of individualism, associational life, and civil liberty are under constant pressure from privacy-denying tendencies toward social egalitarianism, personal activism, and political fundamentalism" (Westin, 1967: 26-7).
It is no doubt interesting to hypothesize that the way the balance between privacy and community obligations and duties is struck within different democratic societies will vary according to different cultural traditions. The belief in privacy is closely related to wider attitudes about participation in public affairs and about trust in the authority of governmental agencies; these questions have attracted considerable attention from students of comparative politics (e.g. Almond & Verba, 1965, 1980), as well as from more anthropological perspectives on social and cultural history (Moore, 1984). Unfortunately, we have little systematic cross-national survey evidence about attitudes to privacy with which to investigate the nature and influence of wider cultural attributes. Most of this argumentation tends, therefore, to rest on anecdotal and impressionistic evidence: "the Englishman's home is his castle" and so on.
The only major comparative evidence by which we might examine the relationship between attitudes toward privacy and political culture are recent opinion polls in Canada and the United States, sponsored by Equifax Inc. and conducted by the Louis Harris firm under the direction of Alan Westin (Harris-Westin, 1990; Harris-Westin, 1992a, 1992b, 1994). Traditional comparative scholarship about Canada and the United States (e.g. Lipset, 1990) has suggested a profound difference in the values of the two countries that explain significant variations in institutional and policy development. The United States is arguably founded on individualist principles, creating a greater distrust in political and administrative elites and in the institutions in which they operate. Canadians are supposed to be more communitarian in outlook, creating a greater deference to authority and trust in public institutions.
Hypothetically, these supposed cultural orientations might be reflected in a greater respect for privacy rights in the US than in Canada. Westin tested this hypothesis in the 1992 survey, and discovered that indeed distrust is highly correlated with concern for privacy in both Canada and the United States. However, "in striking contrast to the traditional wisdom that Canadians are more trustful of government and political processes ... Canadians have a higher combined distrust score than do Americans" (Harris-Westin, 1992b: IX). A follow-up survey in 1994 found the same privacy dynamic at work, with "high and still rising distrust remaining the most significant explanation for still rising privacy concerns" (Harris-Westin, 1994: VI).
We need a systematic cross-national study on privacy which incorporates analysis from Western Europe as well, through which more refined analysis can be conducted about the relationship between privacy and other cultural variables. However, individual polls on privacy in other countries (such as Germany, Britain, Sweden) suggest superficially that populations everywhere have high, and increasing, levels of concern about privacy which seem to be mainly driven by fears of new technology, and distrust of public and private institutions to use that technology with sufficient respect for the civil liberties of the individual. That distrust may be rooted in different historical experience; but it appears to be pervasive and strong (Bennett, 1992: 37-43).
"Political culture" is also hypothetically an explanation for the different approaches to the implementation and enforcement of privacy law. The major variation in the way that Western societies have responded to the privacy problem has been in the institutional oversight mechanisms established to implement and enforce the fair information principles (Burkert, 1982; Bennett, 1988b). To some extent the absence of a data protection agency in the United States can be explained by a typical American aversion to paternalism; Americans are supposed to be able to define their own privacy interests and take steps to protect them through the courts. Indeed, the American penchant for litigation and the consequent proliferation of case law has been defended as a unique approach to privacy, that in sum adds up to a level of "protection" similar to that in societies that have adopted a more comprehensive and anticipatory approach to the problem (Aldrich, 1982; Plesser, 1991).
Marc Rotenberg (1991), however, does not accept this view, arguing that American privacy policy is incoherent and directionless. The sum total of US efforts has been a lot of laws, but inadequate protection. The bewildering collection of federal and state laws can only effectively be used by those with the money to pursue litigation (Smith, 1992). Rotenberg is unpersuaded by arguments about the need for an exceptional and distinctly American approach to the problem that supposedly contrasts with a more interventionist culture in "Europe" that places a protective institution between the individual and the record-keeping organization in the form of the British Data Protection Registrar, or the German Data Protection Commissioner, or the French Commission Nationale de l'Informatique et des Libertes, and so on. He, and others (e.g. Flaherty 1984b), do not accept the argument that such an agency would be an unnecessary and alien import, inconsistent with the American administrative system, even though support for such an agency in public opinion polls is distinctly ambivalent (Harris-Westin, 1990: 104).
Cross-national differences or similarities in the extent of concern about privacy, or in the way that policy outcomes have been shaped, are difficult to attribute in any definitive way to variations in that amorphous concept "political culture." More interesting relationships have been found within the survey research within the United States. In particular, concern for the protection of privacy has been incorporated into a tradition of studies that, since the 1950s, has sought to understand the patterns of support for civil liberties and civil rights among elites and masses. The tradition probably begins with the controversial studies by Stouffer (1955) that claimed to demonstrate a greater respect for civil liberties among elites than among the mass public. The sociological roots of "toleration" have been of subsequent interest to political scientists working within the behavioral tradition (e.g. Sullivan et al. 1982; McColsky & Brill, 1983).
American attitudes about rights to privacy tend to be difficult to dissociate from attitudes about the gamut of civil liberties inherent in the US Bill of Rights. McColsky and Brill's work (1983: 171-231) demonstrates the most interest in privacy issues. Employing data from the 1979 Harris-Westin privacy survey, they address a range of issues about 4th Amendment "search and seizure" questions. Some of the findings are ambiguous. Most Americans, especially elites, are preponderantly opposed to police searches without a court order. On the other hand, they believe (by a two-to-one-margin) that police officers who stop a car for a traffic violation should also be allowed to search the car if they suspect it contains drugs or stolen goods (1983: 185).
Similar inconsistencies are apparent in the findings about wiretapping, which as a general practice is apparently perceived as unfair by most Americans. But they do not seem to regard it as improper to eavesdrop (especially under court order) on the activities of meetings of known criminals, nor to protect the nation against "radicals." The generalization from these and other privacy-related questions is that "popular support for civil liberties tends to be greater when the norms are clear and well established and weaker or more uncertain when the norms are newly emerging, still unfamiliar to many members of the mass public, and not yet firmly fixed in the body of constitutional principles endorsed by the courts" (1983: 185).
While there remains some support in Mcloskey and Brill's study that "community leaders" are more likely to show more tolerance across the range of civil liberties issues (including privacy), there is little evidence that this has much to do with political partisanship or ideology: "at every level of economic liberalism-conservatism, elites score higher on the civil liberties scale than do their demographically matched counterparts in the general public" (1983: 255).
This seems also to be true with respect to attitudes about privacy protection. In the 1990 Harris-Westin poll, it was discovered that "political philosophy does not seem to be much of a factor in peoples' general concerns about privacy" (Harris-Westin, 1990: 1). The same seems to hold in Canada (Harris-Westin, 1992b: VII). The only systematic relationship seems to be that those who self-describe themselves as "moderate" or "middle-of-the-road" tend to be less concerned about privacy than those identifying as "conservatives" or "liberals" (Dutton & Meadow, 1985: 39). Support for the issue also tends to be higher among Blacks and Hispanics in the United States than among whites (Dutton & Meadow, 1985: 38).
The issue seems to be sufficiently broad to embrace the concerns of those over the range of ideological viewpoints: from the civil libertarians who want controls on overzealous law enforcement, to the conservative anti-tax group that is seeking to keep the IRS and their equivalents in check. James Rule asks us to keep in mind that "no one is likely to come out against privacy. But a close look at the clamor for more of it suggests that its proponents do not all have the same thing in mind" (Rule et al, 1980: 135).
There are compelling reasons to believe that the modern issue of information privacy has developed without particular regard to distinctive cultural traditions nor to the influence of different ideological orientations. On the one hand, the technology itself exerts a powerful force for cross-national convergence. What is more, the definition of the privacy problem and the various responses (through data protection or information privacy laws) has much to do with the cross-fertilization of ideas through a transnational community of privacy experts. Lessons were drawn about how best to articulate and respond to the problem. This ultimately led to harmonization efforts by international organizations. The different national concerns for privacy obviously have their source in different historical experiences. Overall, however, it can be argued that the forces for convergence have been overwhelming, and have led to similar concerns, and somewhat similar policy responses.
IV. PRIVACY AND THE POLITICS OF SURVEILLANCE
Some scholars have argued that the contemporary problem confronting advanced industrial states can only be imperfectly addressed and resolved if it is defined in terms of "privacy." Rather the problem is surveillance, or excessive and illegitimate surveillance. "Rather late in the day," David Lyon argues (1994: 219), "sociology started to recognise surveillance as a central dimension of modernity, an institution in its own right, not reducible to capitalism, the nation-state or even bureaucracy."
Within political science, however, the analysis of surveillance has surfaced in relation to the description and critique of authoritarian or totalitarian regimes. This critique spans the centuries, from the use of "spies" within Imperial Rome, to the systematic monitoring of citizen behaviour within Stalinist and Fascist systems (Westin, 1967). To a certain degree, the contemporary interest in privacy among scholars like Edward Shils and Alan Westin was motivated by a desire to build institutional and cultural barriers against the comprehensive monitoring of private life that appeared (during the Cold War years) as a necessary condition for the continuation of totalitarian regimes.
Since this time, however, questions have been raised about the creeping and more subtle forms of surveillance within liberal democratic states. Several questions have been addressed. What is the nature of contemporary surveillance using new information technologies, and to what extent is it different from the practices of the past? What explains the rise of "surveillance societies"? Is it due to an inexorable extension of Weberian bureaucratic rationality? Does it flow from the deterministic logic of technological application? Or is it more rooted in the demands of the capitalist mode of production? Maybe all of these. Perhaps Foucault's point about the ubiquitous and "everyday" nature of power relations, in which individuals unwittingly subscribe to their own surveillance within the "panopticon" provides the central, all-encompassing insight? Is surveillance necessarily a bad thing? At what point does it become excessive and harmful?
From the perspective of those interested in understanding and curtailing social control, the formulation of the privacy problem in terms of striking the right "balance" between privacy and organizational "demands" for personal information hardly addresses these wider questions. The liberal political theory that underpins the "fair information practices" places an excessive faith in procedural and individual remedies to excessive intrusions. Thus privacy and data protection laws can only have a marginal impact on the development of surveillance societies; some would contend that they serve to legitimize new personal information systems and thus extend social control.
The work that is normally cited as the best critique of the theory of information privacy as articulated by Westin is The Politics of Privacy (1980) by James Rule and his colleagues. Privacy and data protection laws are all well and good, but they frame the problem in too narrow a fashion. The argument is that public policies that seek to "balance" privacy rights with organizational demands for information may produce a fairer and more efficient use and management of personal data, but they cannot control the voracious and inherent appetite of all bureaucratic institutions for more and more information on individuals. They cannot halt surveillance, in other words. On the contrary, there are persuasive cases where the enactment of data protection law has been used to legitimate the introduction of new surveillance systems.
The essential problem for Rule then is the inherent tendency of bureaucratic organizations to want to collect and store more and more increasingly detailed personal information. This dynamic of complex organizations has its roots in the 18th century, and the move towards rationalization and control of resources that accompanied industrialization (Beniger, 1986). Thus the "solution" to increasing surveillance can only come from the cultivation of a looser, less discriminating and less efficient relationship between organizations and their clientele.
The arguments of Westin on the one hand, and Rule on the other, have often been posited as diametrically opposed political theories of privacy and surveillance. The distinction between the two sets of literature should not be exaggerated. The difference stems more from the starting-point. For Westin, the problem is privacy and how the institutions of a liberal society might cope with the most dangerous and intrusive threats from new technologies. For Rule (and other sociologists), the starting point is an interest in the changing impact and nature of social control and disciplinary practice. The processing of personal data by private and public institutions is, from this more critical perspective, a way to shed light upon broader social and technological forces.
The idea that advanced industrial societies are creeping inexorably toward an unacceptable level of surveillance has influenced writers from a number of disciplinary and national backgrounds. David Flaherty, a Canadian scholar of legal history, ended up calling his comparative analysis of the operation of data protection laws in Germany, Sweden, the United States, France and Canada, Protecting Privacy in Surveillance Societies. He begins: "The central theme of this volume is that individuals in the Western world are increasingly subject to surveillance through the use of databases in the public and private sectors, and that these developments have negative implications for the quality of life in our societies and for the protection of human rights" (1989: 1).
Flaherty goes on to demonstrate how countries that have established data protection agencies (like Germany and Sweden) have a better chance of stemming the tide than do countries like the United States, whose privacy protection regimes rely solely on the individual assertion of privacy rights through the courts, and on weak oversight mechanisms. But his overall conclusion is skeptical: "How will data protection authorities look by the year 2000? There is a real chance that they will be looked back upon as a rather quaint, failed effort to cope with an overpowering technological tide rather then as a fruitful, successful exercise in promoting the coexistence of competing human and social values" (1989: 406). Echoing Rule's analysis, he suggests that "at present, data protection agencies are in many ways functioning as legitimators of new technology. For the most part, their licensing and advisory functions have not prevented the introduction of threatening new technologies, such as machine-readable identity cards or innumerable forms of enhanced data banks; they act rather as shapers of marginal changes in the operating rules for such instruments of public surveillance" (1989: 384).
The rise of the "surveillance society" (to use Flaherty's phrase) has been traced in a wide-range of less scholarly books from a number of countries that has claimed the presence or imminence of the omniscient and all-powerful state. Orwellian metaphors and imagery are naturally prolific, even though "1984" came and went without any palpable change in the attention paid to privacy questions.
What is significant is the cumulative impact of this literature in several countries. The American literature probably begins with The Naked Society (1964) by Vance Packard and The Privacy Invaders (1964) by Myron Brenton. And continually over the last thirty years publishers have been attracted by this more polemical genre. An incomplete list of the more popular American privacy literature is as follows:
Edward Long, The Intruders: The Invasion of Privacy by Government and Industry (1967)
Jerry Rosenberg, The Death of Privacy (1969)
Arthur Miller, The Assault on Privacy (1971)
John Curtis Raines, Attack on Privacy (1974)
Arieh Neier. Dossier: The Secret Files They Keep on You (1975)
Alan Lemond and Ron Fry, No Place to Hide (1975)
Lester Sobel ed. War on Privacy (1976)
Robert Ellis Smith, Privacy: How to Protect What's Left of It (1979)
David Burnham, The Rise of the Computer State (1980)
David Linowes, Privacy in America: Is your Private Life in the Public eye? (1989)
Jeff Rothfeder, Privacy for Sale (1992)
I have grouped together here a literature that has been written from a diversity of perspectives, ranging from the journalism of David Burnham and Jeff Rothfeder to the civil libertarian approach of Arieh Neier, to the insights of David Linowes (the former chair of the US Privacy Protection Study Commission). The literature also encompasses a shifting concern about emerging technologies, from the concern for "snooping devices" in the 1960s, to the sophisticated trade in personal information revealed in Privacy for Sale.
A parallel, though smaller, literature has also emerged in Britain with a similar tone and emphasis:
Donald Madgwick, Privacy under Attack (1968)
Malcolm Warner & Michael Stone, The Data Bank Society (1970)
Anthony A. Thompson, Big Brother in Britain Today (1970)
Donald Madgwick & Tony Smythe, The Invasion of Privacy (1974)
Paul Sieghart, Privacy and Computers (1976)
Patricia Hewitt, Privacy: The Information Gatherers (1977)
Ian Will, The Big Brother Society (1983)
Duncan Campbell & Steve Connor, On the Record: Surveillance, Computers and Privacy (1986)
Maybe the importance of this literature, however, lies in its cumulative impact and message. A steady flow of horror stories about the intrusive nature of modern technology, about the abuse and misuse of personal data, and about the size and interconnectedness of contemporary information systems has probably had a steady impact on public and political consciousness (Smith, 1993). Moreover, many of these stories are then picked up by the print and TV media. Big Brother imagery, together with accounts of how the powerless can be denied rights and services through the wrongful collection, use and disclosure of personal data certainly make good copy. The media's role in promoting the privacy value, however, is ambivalent; stricter privacy rules would also curtail the prying into the private lives of public figures, and may impede more serious investigative journalism. This tension is most notable in Britain, where "privacy" legislation is now understood to be directed at media intrusions, and has its own separate dynamic as an area of public policy distinct from the question of "data protection" (the European nomenclature, translated from the German Datenschutz).
In general, however, the more journalistic literature generally provides a quite crude picture of the overall nature and effects of surveillance. Its purpose is to draw attention to the complexity and scale of the problem, and to the weakness of existing safeguards. More empirical studies within individual sectors have more successfully exposed the operation and effects of surveillance practices.
An example would be Rule's early comparative analysis of information systems in Britain and the United States (1974). Based on three British, and two American case studies of surveillance systems in both public and private sectors, Rule attempted to assess empirically the changes in surveillance capacity and the attendant implications for social control. One general conclusion was that the overall surveillance potential had grown with the advent of computerized record-keeping in both countries. On the other hand, surveillance systems at that time were limited in terms of the size of the files in the system, centralization, the speed of flow between points in the system, and the number of contact points between the system and the individual (1974: 37-40).
As technology has become smaller, less expensive, and more decentralized, analysts have reached rather different conclusions. Gary Marx's (1988) analysis of undercover police surveillance is a case in point. Marx demonstrates how incremental changes in technology, social values and the law have encouraged covert and deceptive police techniques with a variety of intended and unintended consequences. He demonstrates how all covert surveillance has the tendency to blur the distinction between law enforcement and the lawless activities it is supposed to curtail.
The range of new surveillance practices that Marx discusses allows him to suggest some more general characteristics of these new forms of social control:
1) The new surveillance transcends distance, darkness, and physical barriers; 2) It transcends time; its records can be stored, retrieved, combined, analyzed, and communicated; 3) It has low visibility, or is invisible; 4) It is often involuntary; 5) Prevention is a major concern; 6) It is capital rather than labor-intensive; 7) It involves decentralized self-policing; 8) It triggers a shift from targeting a specific suspect to categorical suspicion of everyone; 9) It is more intensive -- probing beneath surfaces, discovering previously inaccessible information; 10) It is more extensive -- covering not only deeper, but larger, areas. "The awesome power of the new surveillance," Marx summarizes, "lies partly in the paradoxical, never-before-possible combination of decentralized and centralized forms" (1988: 217-9). David Lyon (1992) provides some perceptive reflections on the implications of Marx's analysis for the "maximum security society."
The decentralized, unplanned, incoherent nature of modern surveillance is echoed in other works written in the 1980s. Joseph Eaton's analysis of identity cards demonstrated how, on a de facto basis, Americans already have a national identification card system: "Some people think that the absence of a formal national identification (ID) system is one of the remaining bastions of our right to be let alone. But on a de facto basis, it has already been breached. Other than children, the mentally retarded, and the deceased, how many Americans are without either a social security card, a driver's license, a credit card, or all three of these documents?" (1986: 1-2). He goes on to argue that a truly comprehensive national identity card would provide better protection against fraud, misuse and the invasion of privacy.
The analysis (though not the prescription) is echoed in the excellent 1986 report from the US Office of Technology Assessment, Electronic Record Systems and Individual Privacy. It begins: "the widespread use of computerized databases, electronic record searches and matches, and computer networking is leading rapidly to the creation of a de facto national database containing personal information on most Americans. And use of the social security number as a de facto electronic national identifier facilitates the development of this data base" (1986: 3). The point is effectively demonstrated through a careful analysis of the uses of new techniques for the analysis of personal data: computer matching, computer-assisted front-end verification, and computer profiling.
A new word, dataveillance, has been coined to describe these new forms of surveillance that are facilitated, not by direct visual or audio monitoring, but by the manipulation of personal data. Roger Clarke, one of several Australian privacy analysts, has contended that the "Big Brother" scenario has not arrived because it is unnecessary: "ubiquitous two-way television a la 1984 has not arrived even though it is readily deliverable. It is unnecessary because dataveillance is technically and economically superior" (1989: 499).
There is a wide, and imperfectly understood, range of practices for the analysis of personal data currently used by modern governments. Dataveillance practices vary along five different dimensions: 1) whether personal or mass dataveillance is being conducted; the former involves the analysis of the records of individuals who have already attracted attention, the latter begins with no a priori knowledge of the subjects who may warrant attention: 2) whether the dataveillance is internal or external to the agency that initially collected the data; 3) whether the analysis is upfront or post-facto, that is whether the check is made before or after an individual receives a government benefit of service; 4) whether the analysis is conducted on a single variable, or a multiple number of variables (such as when profiling occurs); and 5) whether the practices have a negative or positive impact on individuals (Marx & Reichman, 1984; Clarke, 1989; Bennett, 1995).
Another Australian writer, Simon Davies, the Director-General of Privacy International, has given some serious thought to the stages through which surveillance systems pass: "As a society becomes larger and more complex, as its links with other nations grow, and as its technological capacity increases, it is normal for it to creep up the surveillance scale" (Davies, 1992: 18). Restricted (Zone 1) Surveillance would only exist in the minimalist "nightwatchman" state. Conditional (Zone 2) Surveillance only exists after adequate debate and the introduction of appropriate safeguards. Routine (Zone 3) Surveillance exists in three principal areas, law enforcement, taxation and government benefits. Mass (Zone 4) Surveillance is a zone of "enforced, interactive and punitive surveillance" in which "most, if not all, aspects of people's movements, transactions, interactions and associations" are monitored. Total (Zone 5) Surveillance occurs when "people show an Orwellian willingness to support government control" (1992: 19-20). Davies argues that many developed countries, including Australia, have been rapidly creeping up to the fourth level of surveillance.
The initial work on surveillance has largely focussed on the activities of the state. Hence, in the works of Rule, Marx, Clarke, Davies, and in the less scholarly work, there is at least an implicit assumption that the explanation for the rise in surveillance is chiefly located in the inherent (maybe pathological) tendencies of bureaucratic organization to want to collect, organize, manipulate and control information. When focussed on the state, the roots of the problem tend to reside in the Weberian theory of rationalization, rather than in Marxist or Foucauldian perspectives.
Two more recent works, which have directed their attention as much to private sector practices, see rather different trends at work. Oscar Gandy's The Panoptic Sort (1992) draws upon a diversity of traditions to try to understand the implications for social control of new and sophisticated practices for the collection, classification, and manipulation of personal information by both public and private sectors. Marx, Ellul, Giddens, Weber, Foucault all contribute to an understanding of the system of disciplinary surveillance that continually seeks to identify, classify and evaluate individuals according to ever more refined and discriminating forms of personal data: "the panoptic sort is a difference machine that sorts individuals into categories and classes on the basis of routine measurements. It is a discriminatory technique that allocates options and opportunities on the basis of those measures and the administrative models that they inform" (1992: 15). His analysis leads him to the conclusion that real consumer choice can only be implemented through "Opt-In" rather than "Opt-Out" provisions.
David Lyon (1994) still employs more visual imagery to address much the same questions about surveillance in The Electronic Eye. He also contends that surveillance cannot be reduced to one social or political process, drawing inspiration from much the same literature as does Gandy. Whereas Gandy relies on contemporary empirical analysis of the surveillance practices of modern corporate and bureaucratic organizations, Lyon adopts a more historical approach to link surveillance to theories of modernity, and to speculate on the possibilities and implications of a more communitarian post-modern condition, as a way to avoid the dystopic visions of both Orwell and Foucault. In this light, surveillance may have positive, as well as negative, ramifications.
The sociological analysis of surveillance systems tries to locate the profound and macroscopic social trends from the temporal or idiosyncratic. But there are inherent dangers in drawing universal judgements from North American experience. The nature and extent of surveillance probably varies across nations, across economic sectors, across bureaucratic institutions. There is a valuable path of inquiry (at a more middle-range theoretical level) which attempts to locate and explain this observed variation. Political scientists have a potential contribution here.
For instance, the sharing of personal data between the agencies of the democratic state for whatever social or political purpose is hypothetically determined by the wider structural configuration of different states, even though the same technological, economic and bureaucratic forces may be at work. For example, it is no accident that computer matching (the comparison of personal data from different files to expose instances of waste, fraud and abuse) should have surfaced first in the United States. Here, public policy is administered through a loose collection of bureaus with very weak vertical lines of authority. This encourages horizontal linkages across government departments thus facilitating the sharing of data. In Britain, by contrast, where these practices are not so widely used, strong lines of hierarchical integration, bolstered by a departmental "esprit de corps" foster secrecy both vis-a-vis the general public, and in relation to other civil service departments. Strong institutional jealousies militate against the sharing of personal data for dataveillance purposes (Bennett, 1995).
Thus the new surveillance may be determined by, and may impact, the internal structural organization of state institutions. It may also, of course, influence the relationship between public and private sector entities. It is becoming apparent that, for example, the files of credit reference agencies are sometimes used to check the eligibility for social services. Private financial institutions are increasingly being considered for the development of "smart cards" and for the dispensing of government benefits. Where do these forms of surveillance leave the distinction between the "public" and "private"?
These and other trends lead Philip Agre (1994) to the conclusion that a "capture" model is just as evocative as a "surveillance model" to represent the new commodification of personal information. This model is built upon linguistic rather than visual metaphors and has its roots in the disciplinary practices of applied computing rather than in the historical experiences of the "surveillance state."
In summary, the literature on surveillance leaves us with the overwhelming message that the quantity and quality of surveillance have changed. The volume of data collected and stored by both public and private organizations has facilitated a range of new practices that have developed incrementally and without much public attention or opposition. By the same token, the surveillance society is not one yet overseen by an omniscient "Big Brother." The formal or legal establishment of a comprehensive national identity system has nowhere been necessary. That system has developed through the uncontrolled decisions of thousands of decentralized public and private organizations, all making supposedly rational decisions that one more incremental invasion of privacy is a price worth paying for greater efficiency and/or profit.
V. PRIVACY, DECISION-MAKING AND THE IMPACT OF NEW INFORMATION TECHNOLOGIES
Westin's initial work also anticipated later debates about the relationship between technology and politics. The analysis of any attempt to promote more privacy in advanced industrial societies immediately raises more profound issues about the deterministic or convergent impact of technology, and the relative autonomy of state policy-makers to control or regulate its more intrusive and dangerous effects. Political scientists have debated such questions at length in relation to the control of nuclear technologies, industrial pollution, television, and so on. Again, however, these questions have rarely been directly addressed with respect to privacy.
That is not to say that the commentary on the privacy implications of the new technology practices that have arisen since the 1960s has not been underpinned by some implicit assumptions about the relationship between technological forces, and organizational or individual choice. The analysis of technologies like computer matching, computer profiling, the development of smart cards, new telephone services (such as call-display), integrated networks, electronic mail, genetic databanks, the "Clipper Chip" and so on, is frequently accompanied by reflections about the nature of the "technological imperative." In this context, the promotion of greater privacy is directly dependent on the ability of individual decision-makers to control these wider structural forces. The debate then raises the age-old conflict between voluntarism and determinism, and the issue of whether outcomes are shaped by structure or by human agency.
Theories about the relationship between technology and politics generally come in three broad categories. The first sees technology as an autonomous or deterministic force. Once technologies are set in motion, they tend to follow their own course independent of human direction. The theory in its purest form is normally associated with the writings of the French sociologist, Jacques Ellul (1964). But it has always been a powerful theme in political and social thought (Winner, 1977). At the other end of the spectrum we find a range of writings that believe that technology cannot be understood outside its social and political context, and therefore can never be regarded as an independent force. On the one hand, it may be shaped by the conscious and autonomous decisions of political agents. On the other, it may be shaped by existent organizational norms or standard operating procedures.
A third, more centrist body of literature, regards this as one big chicken-egg, or cart-before-the-horse debate. This middle position sees outcomes shaped by a complex dynamic interaction between new technologies and existing political and social practices. Here, the outcomes of information technology usage will reflect "the interactive relationship between technological developments, political decisions and existent organizational norms and practices. Technology sets boundaries to achievements, but within those limits, human choice and conflict have considerable latitude (Bennett, 1991: 64). Or as Ithiel de Sola Pool put it: "Technology shapes the structure of the battle but not every outcome" (Pool, 1983: 251). Westin (1980) uses a different analogy to make the same point, seeing technology as a kind of "projectile" propelled into organizations, shaped by and shaping the existing web of rules and practices.
These theoretical positions rest heavily, of course, on prior conceptualization of the word "technology." It is now evident that information technology includes substantially more than the basic hardware and software that is commercially available. James Danziger and his colleagues, in their study of the use of computers in local government, talk of a "computer package" to denote an interdependence of people, equipment and machines (1982: 4-5). Thus the significant unit of analysis is the "technology practice," or the behaviors, rules, conditions etc. that surround the hardware and the software.
Another argument that combines the determinist and bureaucratic control models is that of "function creep." A kind of organic growth is almost universally observed in the implementation of information technology in whatever field of operation. Social and political factors may determine the initial application of a technology. However, once a system is in place, it is rare for its continual growth ever to be reversed. The strategic potential of the technology becomes more obvious to those who use it, but only after it has been implemented for some time. Winner stated the principle as follows: "once underway, the technological reconstruction of the world tends to continue" (1977: 208).
Those who would reject the determinist argument would probably view the privacy problem as stemming from the inherent pathologies of bureaucratic organization. Thus the problem is the acquisition and aggrandizement of power. These assumptions underpin, to some extent, talk about "surveillance states." Here is Michael Stone and Malcolm Warner, writing in 1969: "The computer has given bureaucracy the power of omniscience, if not omnipotence, by putting into its hands the power to know" (1969: 260). US Senator Sam Ervin, the architect of the 1974 US Privacy Act, claimed that "officials at every level of our national life who make decisions about people for limited purposes seem possessed by a desire to know the "total man" by gathering every possible bit of information about him" (1971: 138). Robert J. Gallati, former Director of New York State's Identification and Intelligence System, was more direct: "If we wish to be scientific, it is obvious that we must have information -- lots and lots of data about oodles and oodles of persons and things" (Gallati, 1971: 42-3).
The perspective of those who root the essential privacy problem in the inherent corrupting influence of power is also more likely to be influenced by historical experiences.
The technology is more likely to be regarded as a mere "tool" in states with more vivid memories of the abuse of political power. This is certainly a central motivation behind data protection law in continental Europe. A 1984 conference of data protection experts concluded that "one of the prime motives for the creation of data protection laws in continental Europe is the prevention of the recurrence of the experiences in the 1930s and 1940s with Nazi and fascist regimes" (Flaherty, 1984a: 5).
Those who have conducted any empirical research into the development and application of personal information systems tend, however, to adopt a more centrist position. Here is David Flaherty, writing in 1989: "Civil servants seek data on individuals to design and evaluate programs, to augment their prestige and power, and, as a product of a supposed technological imperative, to enable them to use the latest hardware and software" (1989: 13). Arthur Miller (1971: 21) articulates a "Parkinsons Law" of privacy invasion, similar to the notion of "function creep": "Technological improvements in information-handling capability have been followed by a tendency to engage in more extensive manipulation and analysis of recorded data."
The development and application of personal identification numbers can also be effectively understood as a process of "function creep." The pattern seems to be pretty consistent everywhere. Numbers such as the American Social Security Number, the Canadian Social Insurance Number (SIN), the Australian Tax File Number, and the British National Insurance Number, were originally introduced solely for the purpose of administering social benefits and unemployment insurance programs. Their utility as identifiers for other purposes was then recognized by other public institutions. Gradually they have been used for a variety of unrelated purposes to the extent that they have been regarded as de facto universal identifiers. The only country that has attempted to regulate their usage is Canada; in 1988 the Canadian government, in response to pressure from privacy advocates, announced the phasing out of all but seven uses of the SIN.
Empirical investigations into the development and effects of specific personal information systems also provide some insights into the processes of IT acquisition and development. The earliest, and still the most comprehensive, is Alan Westin and Michael Baker's project for the National Academy of Sciences, Databanks in a Free Society. Based on site visits to 55 government, commercial and non-profit organizations, the findings of this project indicated that the "content of computerized records about individuals has not been increased in scope compared to what was collected in their manual counterparts in the precomputer era" (1972: 244). They concluded that "pre-computer rules have not been altered in computerizing organizations; rather, customary practices have been reproduced with almost mirrorlike fidelity" (1972: 253).
This study was initially cited as evidence that the computer was not the villain, and that attention to the privacy issue had been exaggerated. But the Westin and Baker findings were based on 1970-72 data, when the cost of computerization was far greater than it is today (Regan, 1981: 78-9). Westin and Baker did, however, make some perceptive forecasts about the development of these "increasingly powerful and flexible tools" that will need to be shaped by the values and standards of democratic society (1972: 334).
The later analysis of criminal justice information systems conducted by Ken Laudon for the Office of Technology Assessment (US, OTA, 1982), revealed that some of Westin and Baker's warnings had not been heeded. This study found that the use of criminal history records for purposes other than those for which they were collected was extensive and growing. It was found that criminal record checks were conducted for employment and licensing purposes, especially for applicants to positions entrusted with the care of persons who are likely to be vulnerable (especially children, the elderly, the mentally handicapped). The incremental use of a technology in place for seemingly legitimate social purposes, by the early 1980s, had amounted to a surveillance system over which there were few controls and about which there was much concern (Laudon, 1986).
Donald Marchand's The Politics of Privacy, Computers and Criminal Justice Records (1980), also found that the use of criminal justice records outside the traditional confines of the criminal justice system was enormous and increasing. Privacy was viewed as one largely symbolic way that the political system may exercise some control over the "technology reception" process: "The recognition of this right, in view of increased organizational recordkeeping and computerization and the adverse effects of these activities on individuals, has meant that privacy is not only a concrete right to be reconciled with competing social goals and rights, but, stated abstractly, also is a political resource" (Marchand, 1980: 88). A careful empirical analysis of the case of criminal justice records in the United States allows Marchand to examine the role that privacy plays within the wider interaction between technological development, organizational norms, and political choice.
Evidence from public opinion polls also suggests a strong, if poorly understood, fear of computer technology. Distrust of technology appears to be closely tied to distrust of institutions (Harris-Westin, 1990: XXII). Around two-thirds of the American public agreed in the 1991 update survey that "if privacy is to be preserved, the use of computers must be sharply restricted in the future" (Harris-Westin, 1992a: 20). Evidence also suggests, however, that those who know how to use computers are less afraid of the "1984" scenario that those who do not (Dutton & Meadow, 1985). Business elites are also far less likely to regard computers as a threat to privacy than are the general public (Harris-Westin, 1990: 83).
It is, however, a mistake to believe that information technology is necessarily antithetical to privacy interests. Recent advances in the field of public-key encryption have demonstrated that some technologies can also be privacy friendly (Katz, 1987: 84; Chaum, 1988; 1992). David Chaum's basic point is that common identifiers such as social security numbers inevitably involve a trade-off between security and individual liberties. But advances in microelectronics, and in the development of secure "digital signatures" allow people to adopt a different (but verifiable) identifier to every organization they do business with. Theoretically, this would make "dossiers" impossible and unnecessary. Organizations would also benefit from increased security and lower record-keeping costs.
The obstacles to the application of encryption technology are probably political, economic and psychological rather than technical. However, encryption can be a two-edged sword. The US Government's recent decision to pursue the "Clipper Chip" as a national standard for encryption has been severely criticised by privacy experts for its ability to enhance the government's capabilities for electronic surveillance. It has also met with stiff opposition from the private sector which sees it as discouraging the development and export of new cryptographic technology (Barlow, 1994).
Thus encryption technology is also susceptible to the process of "function creep." The bureaucratic imperative to make the most efficient use of public resources (including personal information) has been overwhelming within recent conditions of high deficits, increased levels of criminality and neo-conservative economics. Thus, if a technology is available, the likelihood is that it will be used for one of these wider social and governmental purposes. That, I would argue, is the history of computer matching, of the development of identification numbers, and of the application of new telecommunications services (such as call-display telephones). I would also predict that when the history of the uses of genetic information is written, similar forces will be shown to be at work.
Privacy may also be seen by political scientists as a regulatory policy, the development of which may tell us something interesting about how different states manage technological change. In this light, the issue raises a set of fairly traditional questions for the policy analyst: How did privacy reach the political agendas of different states? How was the issue articulated by interest groups and policy communities? How was it pursued through the legislative process of different countries? How were policy instruments chosen for its enforcement and implementation? How has privacy (data protection) policy been implemented?
How has its success been evaluated?
The analysis of privacy as a political issue can yield some fascinating insights into the capacities of different countries with different institutional arrangements and policy legacies to define, recognize and respond to, a common problem. It is, therefore, ideally suited to the approaches and techniques of the comparativist. It permits a nice distinction between the "persistent, generic, and transnational from the conditional, particular, and country-specific" (Bennett, 1992: x). It also allows a combination of the analytical approaches of international relations with those of the comparativist.
The number of works that have analyzed the development of privacy policy within an explicitly comparative framework is, however, very limited. Flaherty's (1989) comparison of the implementation of data protection law in the United States, Canada, Sweden, West Germany and France is obviously the most notable. This was complemented in 1992 by my own Regulating Privacy, which attempted to study the formation of data protection policy in the United States, Britain, Sweden and West Germany. I should also mention in this context Priscilla Regan's (1981) Ph.D dissertation on approaches to privacy protection in Britain and the United States, and related articles (1984, 1990). Most of the comparative literature, however, is within the field of comparative law, is more descriptive and normally makes little attempt to understand the law within the broader social, cultural or political contexts of different countries (e.g. Hondius, 1975; Bing, 1978; Schwartz, 1989).
Nevertheless, the privacy literature can be used to draw some wider conclusions about privacy as a regulatory policy and how it has been treated in different advanced industrial states. Dotted here and there appear a number of short, but quite interesting discussions about the politics of privacy and data protection that together add up to some insightful stories about policy-making. Some of those stories were traced in Regulating Privacy. Much more could be done (see, Regan, 1995, forthcoming).
First, how did privacy reach the agenda of advanced industrial states? The analysis of how issues hit agendas, how they become "issues whose time has become" (Kingdon, 1984) has attracted the attention of many political scientists. There are some common patterns. In all countries, to some extent, plans for the computerization of government information systems and for the extension of universal identification numbers caused the most notable concern. Another common variable is the role of expertise; in most countries, it seems, one or two key figures gained a reputation for being the experts on privacy and data protection. A body of knowledge circulated amongst this elite, and was readily available to any governmental decision-making body.
But there are also some interesting variations. In some countries, the issue emerged steadily and quietly as an elite of experts gradually and skillfully articulated the problem and convinced key policy-makers that legislation was necessary; this was largely the pattern in Sweden, for instance. In others, a "triggering mechanism" was necessary, such as the Watergate scandal in the United States, or the uproar over the "Safari" personal information system in France (Flaherty, 1989: 166). In the countries that have legislated most recently (for example Britain, Japan, the Netherlands), it is probable that economic considerations, relating to the trade implications of transborder data flows, were the key motivation.
The role of interest groups has also yet to be properly analyzed. In all states, civil liberties associations played some part in articulating the concerns. More interesting work could focus on the role of professional associations (such as those in the medical, legal and financial fields), and on the different coalitions that it was necessary to build in order to inject data protection policy into different sectors. In this regard, there has been a little analysis of the role that industry played in lobbying for the British Data Protection Act of 1984 (Bennett, 1992: 91). Regan's (1993b) case study of the passage of the Electronic Communications Privacy Act in the United States is also evidence of the importance of the building of coalitions of interest groups (in this instance between key industry associations and the civil liberties lobby) for the passage of privacy legislation.
As attempts to strengthen privacy law in the private sector gain momentum at the international level (through such instruments as the EU Directive on Data Protection) and within North America, important opportunities will arise for the analysis of the political influence of the "data-intensive" industries. These include sectors such as credit reference, insurance, banking and particularly direct marketing. There is already compelling evidence that the latter has been especially active in lobbying against the "informed consent" provisions within the EU Directive (Raab & Bennett, 1994). The politics of this issue may take on a very different character when the resource to be regulated (personal data) is one upon which many organizations rely for their very existence.
By far the most attention in the privacy literature has been paid to the next "stage" in the policy-making process, namely the formal legislative history. Several books and articles give brief accounts of how the privacy or data protection bill became a privacy or data protection law. One especially interesting story was how the American Privacy Act came to be passed without a provision for a Privacy Commission (see Flaherty, 1984b; Regan, 1984; Linowes & Bennett, 1986). The story reflects a typical American process of bargaining and logrolling. Differences in the Senate and House versions of the Privacy Act required last-minute compromise. The Privacy Commission, originally proposed by Senator Ervin, was ultimately removed in favour of an analytical body, the Privacy Protection Study Commission, and oversight by the Office of Management and Budget and by Congress.
David O'Brien (1979) adds an interesting chapter on "Personal Privacy as an Issue of Public Policy" to his otherwise fairly traditional public law analysis of the subject. He attempts to place American federal privacy policy within a framework for understanding the informational flows that are demanded by all information policy (including the Freedom of Information Act) (O'Brien, 1979: 221). Privacy, for O'Brien, is as much a political ideal as a legal right or an existential condition.
Comparative legislative histories are also briefly traced for the countries that Flaherty (1989) studies. Other short accounts appear in several other works on Britain (Campbell & Connor, 1986), on West Germany (Simitis et al., 1981; Bull, 1984), on Australia (Davies, 1992), and so on. The juxtaposition of these stories, I have argued, allows the political scientist to say something more generally about styles of public policy-making (1992: 229-38). In some states, the politics were largely consensual (e.g. in Sweden); in others major battles were fought (such as in Australia and West Germany). In some, the process was short (such as in France); in others it was drawn-out (e.g. in Britain and the Netherlands).
The chief battles in most countries were fought not over the principles of fair information practice, but over the methods and structures through which they would be enforced. A cross-national convergence around these principles can be contrasted with a divergence over the "policy instruments" that different states chose. This allows the political scientist to examine a central hypothesis of the comparative public policy literature, which suggests that states have a basic repertoire of "tools" that they apply in any problem-solving situation (Hood, 1986). I found that there was some support for this hypothesis in the data protection case (Bennett, 1992: 195-200); some states do have some fixed notions about which policy instruments "fit" into the wider institutional setting. On the other hand, the specific powers granted to some data protection agencies were also influenced by political bargaining between elected and unelected officials (Bennett, 1988b).
The experience of the implementation of privacy and data protection law also permits an interesting interrogation of theories and hypotheses drawn from wider political science literature. Regan's (1984: 19) statement about the dilemma of implementation certainly provides a very realistic picture of the difficulties facing policy-makers in this area. With reference to the United States and Britain, she demonstrates how, "when implementation questions are left unresolved in policy design, bureaucratic concerns dominate the implementation stage; yet, when implementation questions are resolved in policy design, bureaucratic concerns dominate the formulation stage." This dilemma arises in this peculiar area of regulatory policy, in which the regulated (i.e. bureaucratic agencies) are also the regulators.
The continuing battles over the development of new personal information systems between bureaucracy and data protection agencies are traced in detail in Flaherty's (1989) volume. The conditions for successful implementation are inductively drawn out at the end of the study. These can be summarized in the following way: 1) a constant review and adaptation of the statutory framework; 2) an expertise in the capabilities of new technology; 3) a separate policy instrument with a permanent, relatively small, dedicated, administratively expert staff whose responsibilities for onerous licensing and registration tasks are minimized; 4) a single privacy advocate at the head of the agency who knows exactly when to use the carrot and when the stick; and 5) a supportive public, media and legislative opinion.
He does not like part time commissioners, or collective committees (such as the Commission Nationale de L'Informatique et Libertes in France). In summary, he concludes that data protection agencies should not be a kind of "miniparliament that seeks to settle the appropriate balance internally ... Emphasis should be on the antisurveillance side of the balance, since the forces allied against privacy, or at least in favor of effective surveillance, are generally so powerful" (1989: 391). Spiros Simitis, the prominent Data Protection Commissioner from the state of Hessen, drew some similar conclusions about the necessary elements of any regulatory system in an influential 1987 article. Schwartz (1992), too, uses knowledge of the German data protection regime to critique the incoherent American response to the problem.
It would be very profitable, however, if someone could apply some of the political science literature on implementation to privacy and data protection law. Flaherty does cite Daniel Mazmanian and Paul Sabatier's (1983) framework for effective policy implementation. Other more contemporary works may yield more rewarding hypotheses. Charles Raab (1993), for example, has contended that some models for implementation analysis rely too much on a top-down approach, in which variables are deductively analyzed as a kind of checklist. This model may not be suitable for the analysis of data protection. Rather, more dynamic approaches would be more sensitive to the process of mutual learning which is, of course, indispensable to any data protection agency and especially to those that have to rely on advisory rather than regulatory strategies.
The final set of questions for the policy analyst have to do with evaluation. Do privacy or data protection laws achieve their stated goals? There has been some attempt by data protection agencies to think about possible performance measures to justify their resources; there was a session at the 1992 Annual meeting of Data Protection Commissioners in Sydney on this very subject.
Policy evaluation techniques have been developed and applied in many other areas of regulatory policy, many of which are admittedly more amenable to more quantitative or objective assessment methods. Nevertheless, political scientists have given careful thought to performance measures, and similar analysis can probably be applied to privacy and data protection too. The initial why, what, how and who questions that any policy evaluation would need to address were outlined in a 1993 paper to the 6th Annual Conference of Privacy Laws and Business (Raab & Bennett, 1993).
The task is a complex one. But it is important to begin to grapple with the essential issue of what constitutes "policy success" in this area, and who should evaluate that success and how. This depends on how we observe, and maybe measure, protection. Among other things, we need to question the assumptions behind instruments like the EU Directive that countries can be ranged on a clear scale, with those with "high" levels of protection at the top, those with "adequate" protection in the middle, and those with "inadequate" or "inequivalent" protection at the bottom.
VII PRIVACY, CAPITALISM AND THE INFORMATION MARKET
The assumption behind most of the literature cited above is that, however framed, privacy is an important democratic right that the state has some obligation to protect through regulatory policy. In some countries, such as the United States and Canada, that obligation generally extends only so far as the personal information held by federal and state or provincial governments. In others, the policy problem is present wherever large quantities of personal data are stored -- in both public and private sectors. But in virtually every advanced industrial country, a privacy (or personal data protection) policy has been fashioned. Privacy is now almost universally seen as a political problem.
Some, however, arguing from economics or public choice assumptions contend that, as privacy is an inherently private good, it can be better protected through individual responses and market-based mechanisms. Those who have argued this case and challenged the conventional wisdom have been lone voices. However, as the scale and complexity of the information economy increases, so more privacy analysts have been looking to market-based mechanisms in the belief that the "information superhighway" of the future will likely be impossible to oversee and regulate.
One key assumption here is that privacy is not a final good, but an intermediate or instrumental value: "Under this approach, people are assumed not to desire or value privacy or prying in themselves but to use these goods as inputs into the production or income of some other broad measure of utility or welfare" (Posner, 1978a: 19). Flowing from this distinction between ends and means, Posner contends that "with regard to ends there is a prima facie case for assigning the property right in a secret that is a byproduct of a socially productive activity to the individual if its compelled disclosure would impair the incentives to engage in that activity; but there is a prima facie case for assigning the property right away from the individual where secrecy would reduce the social product by misleading the people with whom he deals" (1978b: 403).
The economics of personal information are controversial. Richard Posner's initial (1978a, 1978b) exposition of the theory drew much criticism (e.g. Bloustein, 1978). The policy implications of this analysis lead to common law solutions, in which there has never been recognized a right for individuals to conceal discrediting information about themselves. Posner's analysis does, of course, rely on a clear and objective distinction between true and false, or harmful and harmless, information. Others have shown that the nature of the information is not the crucial element, but rather the context in which it is stored. My name and address in the telephone directory is normally harmless; that same information on a list of bad credit risks is capable of doing me considerable harm.
This, and other, critiques (e.g. Stigler, 1980; Gould, 1980) of the theory of information privacy, and the "fair information practice" doctrine that underpins the statutory approach has, however, led a more recent search for marketplace solutions to the privacy problem. These explorations take a variety of forms, and do not necessarily follow the logic presented by Posner. Nevertheless, the space for debate about alternative approaches has opened up.
We see, for instance, a burgeoning debate (especially in North America) about the incentives that business might have to promote "privacy friendly" practices. There is some evidence that industries within the credit, financial, direct-marketing and telecommunications sectors, for instance, have recognized the economic importance of promoting an image of being responsible users of personal information through the implementation of voluntary codes of practice. The development of mail and telephone preference services which allow consumers to "opt-out" of direct marketing lists can also be explained by similar motives. Although, as Jeff Smith demonstrates in his recent empirical analysis of the personal data practices of American business, the extent and nature of that recognition is variable. The process by which businesses develop policies to control their personal data holdings is generally a "wandering and reactive one" (1994: 55).
Changes in the fundamental nature of marketing are also responsible for shifting concerns about personal privacy (Culnan, 1993). Many have noted the trend away from mass-based advertising toward more targeted, direct marketing strategies. Sophisticated "psychographic" techniques, which allow a refined analysis and profiling of consumer attitudes and lifestyles have directed the search for more and more precise information about individual preferences and behaviours (Smith, 1994: 76-7). Direct marketers argue that they can only reduce the nuisance value of "junk-mail" by collecting and sorting more precise and accurate data about individual consumers. Nevertheless, companies in these "information-intensive" sectors are well aware that consumer privacy concerns can disrupt their businesses. Consumers are not, therefore, powerless, although the depth and scope of potential consumer pressure for privacy has yet to be clearly understood.
If there is a business incentive to promote privacy-friendly practices then common codes of conduct might potentially spread throughout the economy in the same way that product and performance standards have been applied in other areas. The Canadian Standards Association has recently negotiated a common Code for the Protection of Personal Information to act as a minimum standard for privacy protection within Canada's private sector. Hypothetically, certification and registration processes can be administered through accredited standards-setting bodies. Privacy would then spread as market pressures, consumer demands, government procurement policies, international forces and so on, would force Canadian business to adopt the privacy code and verify to an independent auditor that it does what it says it does (Bennett, 1995).
Ken Laudon certainly realizes a potential for creeping privacy protection through consumer demand. He has recently produced a coherent critique of privacy law, arguing that the conceptual foundations are outdated due to changes in the nature of the marketing economy, facilitated by changes in the capacity and speed of modern computer and communications technologies. He believes that personal information can be regarded as a form of property, in which the individual has not only a "mere juridical or administrative interest" but a stronger property interest (Laudon, 1994: 66). He contends that "to ensure the protection of privacy beyond 2000, we should consider market-based mechanisms based upon individual ownership of personal information and National Information Markets (NIM) where individuals can receive fair compensation for information about themselves" (1994: 65). The fair information practices regime was developed to deal with the privacy problems inherent in a relatively small number of large-scale mainframe databanks, rather than a fluid, decentralized and networked computing environment. Where one personal information system begins, and another ends is becoming impossible to define.
Lawrence Hunter and James Rule have recently come from a sociology background to argue the benefits of a somewhat different market-based solution to the sale and trading of personal data (Hunter & Rule, 1994). They advocate legislation that would grant every citizen the "rights to commercial exploitation of information about himself or herself" (1994: 4). They would also be guaranteed the option to insist on royalties in exchange for the commercial release of personal data. This would arguably generate a new kind of business -- information brokers who would represent individuals' interests in the treatment of their personal information (1994: 5).
Or perhaps personal information has variable "value" depending on the type of institution that collects it. In this vein, Eli Noam has proposed a multi-tiered approach to privacy protection, in which certain levels of privacy not developed through legislation or constitutional interpretation would be "left to the market to allocate" (Noam, 1994: 35). These conclusions stem from Noam's own work on New York telecommunications privacy.
Each of these authors starts from different ideological and methodological assumptions to perceive a different crisis to come to these very different forms of market solutions. Laudon perceives an inherent weakness in FIPS doctrine; Noam emphasizes a rapidly changing technological environment within communications; Rule and Hunter see a crisis for consumer choice and empowerment. Each also have studied different institutions, in which the record-keeping relationships pose subtly different privacy problems, requiring different solutions.
Moreover, each also proceeds from the observation that information surrendered in the course of applying for a government service, purchasing goods, using computer or communications technology and so on may be appropriated by third parties who normally have not engaged in any direct transaction with the data subject. Third parties (such as list-brokers and their telemarketing clients) are thus classic "free-riders" in the economy. There is a misallocation of the costs and benefits, because these third-parties consume a common property without covering the externalities related to its extraction and use (Lawson, 1995).
Posner and others of the Chicago school would contend that as privacy is only an intermediate good, there are no costs incurred when a third-party appropriates that personal information, unless privacy is necessary to prevent some final harm. Thus the imbalance in the allocation of costs and benefits in the information marketplace is an illusion, because the loss of privacy is not a basic "cost" if individuals have "nothing to hide." This critique would explain why it is a sociologist like Rule, believing that privacy is a fundamental right (a "final good" in economics terms), who has proposed a property-based solution to the exploitation of personal information in the information marketplace.
From a civil libertarian stance, the debate about market forces is incidental to the central issue that privacy should be a "public good." Consumers may have some bargaining power with a direct marketing firm that wants to trade lists; citizens, however, have no bargaining power when faced with a warrant or any other potentially privacy-invasive technique backed up by the sanctions of the state (Rotenberg, 1994).
VIII. THE POLITICAL ECONOMY OF GENETIC INFORMATION
The definition of new technological issues as a potential problem for personal privacy tends to occur according to a clear and predictable pattern. An initial period of concern, accompanied by real horror stories or more hypothetical fears, is then succeeded by a process of study. In the United States, this analysis may be conducted by committees of Congress, or by associated agencies such as the Office of Technology Assessment, or the General Accounting Office. Less typically, it may also be performed by organs of the executive branch (such as the Office of Management and Budget, or the Department of Health and Human Services). In addition a variety of public, or quasi-public, agencies, foundations and think-tanks have periodically decided to fund the research of private scholars. In countries which have a Privacy or Data Protection Commission, this research tends to be coordinated through this office.
This initial period of analysis is then followed by the development of either legislative, or voluntary privacy rules or guidelines. In the United States, Congress may enact some specific legislation, such as the Computer Matching and Privacy Protection Act of 1988, or the Electronic Communications of Privacy Act of 1988, or the Video Protection Act of 1988. In societies with comprehensive data protection laws, the debate tends to concentrate on how the new technology practice can successfully by integrated into existing privacy protections. But everywhere there is a discernible rhythm to the entrance and exit of privacy issues from political agendas. The "issue attention cycles" of privacy concerns such as wiretapping, credit reporting, educational records, computer matching, social security numbers, call display telephones, videosurveillance, medical privacy, workplace monitoring and so on, require further analysis from students of comparative public policy.
With respect to the privacy of genetic information, as Westin shows, the early alarm phase is now giving way to empirical analysis and investigation (Westin, 1994: 68). Typically, the United States seems to be leading the way in scientific discovery and application in this area, and through the funding of the Ethical, Legal and Social Implications (ELSI) program is examining the privacy impacts of genetic technology in a more anticipatory way than has been typical with other privacy-invasive technologies. Societies with institutional mechanisms for privacy protection have begun analyzing the problem under the auspices of these agencies (e.g. Privacy Commissioner of Canada, 1992).
Debates about the collection, use and disclosure of genetic information in the United States have tended to center on some fairly traditional questions about the rights of informational self-determination, control over disclosures, the protection of anonymity and the nature and process of "informed consent" (Powers, 1994). The moral justification for privacy rules in this area then leads to a number of tricky questions concerning the appropriate level and extent of regulation. Should there be special policies and rules for the privacy protection of genetic information, or should genetic data be viewed as no different from any other kind of medical information? What is the appropriate role for the federal government, versus the states? What should be the nature and scope of legal remedy? The search for the correct "balance" between individual and organizational interests constitutes the central, and typical, purpose of policy debate.
In contrast to privacy issues that focus on "records," however, there has been a more active debate about whether there is something intrinsically different about genetic information that demands special attention and regulation. The following characteristics have been noted: 1) the long-term predictive qualities of genetic data; 2) the fact that genetic testing can reveal information about other family or group members besides the individual; 3) the stigmatizing effect of knowledge of genetic disorders; 4) the intimately personal nature of the information and the fact that it defines and constitutes our identities, unlike information that is provided or acquired through some external agency; 5) the immutability of that information; and 6) the fact that it relies on probabilities for whole populations, and therefore may be tragically misleading or incorrect in individual cases. The debate will continue among specialists about the specificity of the rules to govern medical practice in this area.
The preceding analysis about the political economy of privacy does, however, lead me to emphasize that the "problem" to be rectified here stems not so much from the intrinsic nature of genetic information, but from the familiar pathologies and frailties of social and political institutions. The literature analysed above should remind us of the timelessness of these sociological and political questions.
The literature on the political theory of privacy should serve to warn us about the shifting and controversial definition of the "self" in modern society. The collection, use and dispersal of genetic information (the data that define us as individuals) reminds us how important it is to think about the genetic privacy question in terms of larger issues about the boundaries between the "individual" and "society." We might reflect on how the addition of a genetic profile contributes to the proliferation and multiplication of individual identities within the everyday "normalizing gaze of the Panopticon."
Those boundary questions take on a particular significance when the data being collected and used for community purposes are not a record or a number but the essential and intrinsic stamp of our identities. Genetic privacy maybe brings into question more forcefully than any other privacy issue the "fair information principles" doctrine, and the assumption of "balance" upon which it rests. To put the issue in Westin's terms, the choice will be between the position of the "privacy fundamentalists" (which are committed to privacy protection at the expense of other social goals), and that of the "privacy pragmatists" (which are more willing to tolerate trade-offs) (1994: 64). Powers (1994: 82) contends that a "morally acceptable public policy with respect to genetic privacy will be a mixed strategy or one in which elements of both the fundamentalist and pragmatist view are incorporated, depending on the particular balance of competing interests at stake."
From the literature on political culture, the analysis of genetic privacy may draw upon wider findings about the relationship between concern for privacy and trust in different institutions. Public policy for the processing of genetic data should be guided by clear findings from public opinion surveys that some institutions are clearly regarded with less suspicion than others when it comes to treating personal information fairly. Among Americans, it is relevant that hospitals are most likely to be accorded a high or moderate degree of trust in the responsible use of information than virtually any other institution in either public or private sectors (Harris-Westin, 1990: 19).
From the literature on surveillance, we should receive plenty of forewarning against the capabilities of genetic databanks. The potential for the whole range of "dataveillance" techniques is enormous. Given the greater predictive qualities of genetic data, matching and profiling have a more powerful analytical potential. The possibility for social control inherent in such systems will shift the proposed policy solutions away from pragmatic trade-offs toward more fundamentalist goals.
From the literature on information technology and decision-making, we should certainly beware the process of "function creep." This is probably already happening with the use of genetic data, with possibly serious consequences. There is nothing to lead us to suppose that this kind of information will be any less susceptible to organizational pressures for its use for purposes other than those for which it was collected. These administrative pressures might conceivably be stronger in the case of genetic data, given how genetics can reveal much more about an individual than other types of medical testing and evaluation (Powers, 1994: 80).
The issues raised in the literature on regulatory policy-making are also of potential relevance. Americans should remain sensitive to the fact that the same dilemmas about privacy rules for genetic data are being actively debated in all advanced industrial states, many of which have privacy and data protection agencies that can more forcefully articulate the privacy interests. The incentives for international lesson-drawing are likely to be considerable.
The pressures for the harmonization of rules for the transfer of genetic data may also outweigh national interests and approaches. The globalization of privacy problems is always a trend to keep in mind (Regan, 1994). Genetic data (like other forms of information) will know no national or institutional boundaries. There will be technological and institutional pressures for the convergence of public policies on genetic data, just as there are for the harmonization of other privacy rules.
American policy-makers might also reflect on the difficulty of analyzing and responding to this privacy problem in the absence of a specific institutional mechanism, such as a privacy commission. The US response, while more anticipatory than normal, is still fragmented and incoherent (Gellman, 1993). There is a stream of policy analysis and advocacy, and a separate stream of political and legislative activity (Kingdon, 1984). Whether these two coincide and a consistent set of rules for the collection, use and disclosure of genetic information emerges will, on past experience, be a matter of chance rather than design. The American response to the privacy problem has always been susceptible to the vagaries and unpredictabilities of a legislative system more suited to distributing particular private goods to constituents and clienteles, than in fashioning coherent public policy in anticipation of social dangers.
Finally, changes in the nature of the information economy raise issues about the value of genetic information and the potential to regulate its circulation through means other than law. The nature of "informed consent" is of course actively debated in this area, suggesting that individually determined and articulated solutions have a key place in any public policy about genetic privacy. The extent to which genetic data has an economic value, and is therefore amenable to the kinds of solutions advocated by Laudon, Noam and Rule, remains open for debate (see, Regan, 1995, forthcoming).
Thus we should not only ask whether and how genetic information is "different", we should analyse how its use by existing political, administrative and economic institutions can be shaped by democratic values. On a fundamental level, the genetic privacy issues (like all "privacy" issues) is a problem of political and economic power. The problem should lead us to try to understand not only the different quality of genetic data, but why, for example, some employers would want to use genetic information gleaned from hair follicles as an aid to job-screening (Gandy, 1992: 132).
IX. CONCLUSION: POLITICAL ECONOMY AND THE PRIVACY DEBATE
The number of academic political scientists cited in this review has been few. There are two reason for this. First, political science, and especially American political science, tends to gravitate towards institutions, that is agencies with rules, personnel and budgets that can be empirically observed. Given the absence of a privacy or data protection commission in the United States, the issue has floated around a vast and diverse collection of congressional committees and executive agencies, all of which have other and generally more significant responsibilities.
The second reason is that privacy tends to span the traditional subfields within the discipline. It is an issue of political theory, of public policy-making, of political behavior, of public administration, of comparative politics, and of international relations. To study the "politics" of privacy, as I have discovered, requires a certain, though inevitably superficial, familiarity with the conversations at these "separate tables." In conclusion, however, I want to suggest some of the weaknesses of the way we debate the politics of privacy, and suggest ways that political science might invigorate the discourse. The debate has been, for the most part, overly reactive, ahistorical, non-comparative, and descriptive.
First, the task of the privacy advocate or analyst can be likened to that of trying to change the tire of a moving car. Information technologies are being developed
and applied with startling rapidity and creativity. The task of the privacy community has always been to discover these emerging technologies, analyze their impact on privacy and then to fashion the appropriate regulatory response. These debates have generally been reactive rather than anticipatory. Understandably, they have been focussed on the pressing practicalities of the moment.
This is far truer today than in the 1960s. The cumbersome mainframes of thirty years ago were discrete and uncomplicated compared with the perplexing variety of new technologies that contemporary privacy analysts have to get their minds around: smart cards, Caller ID, genetic fingerprinting, computer matching and profiling, encryption technology and so on. The annual reports of the world's privacy commissioners, as well as their annual conferences, discuss an increasing range of different technologies, most of which were never anticipated by national data protection laws nor contemplated as a responsibility for the data protection authority (see, e.g. UK Data Protection Registrar, 1993; Canada Privacy Commissioner, 1993).
By and large, neither our data protection authorities nor the relatively small (but growing) group of international privacy advocates have the resources to anticipate technological developments and research their various impacts.
Second, the analysis of privacy and public policy has been ahistorical. That is not to say that the value of privacy has not been analysed using historical methods. Through the analysis of privacy, some historians have certainly been able to illuminate the social conditions of our ancestors, and demonstrate that privacy is deep-seated and intrinsic human concern (e.g. Flaherty, 1967; Seipp, 1978). Rather, in our focus on the latest technology, the latest law, the latest version of the EU's Directive on Data Protection and so on, we tend not to see the policy issues in historical perspective. If we did, some of us may not be so pessimistic. In just over twenty years, most advanced industrial states have enacted information privacy or data protection laws. The diffusion of this law has taken place largely without any major clamouring from powerful outside interests, and with only rare outbursts of public or political enthusiasm. If we reflect on these historical developments, we may conclude that, even though there are still enormous weaknesses in some of the laws, the issue has come a long way in a relatively short period.
A historical perspective can also help us understand the origins of current proposals. Take, for example, the Draft Directive on data protection from the EU that has been dominating debate for the last three years or so. This Directive is the culmination of a set of processes that were originally set in motion in the early 1970s. It is the latest (maybe the last) instrument of policy convergence, and builds upon previous international agreements that harmonized law on the level of some very general principles. The EU Directive directs a magnifying glass on a variety of more specific divergences in European data protection. This Directive would not have been possible without earlier harmonization efforts and the structures and processes that were set in place in the 1970s. Unless we observe these developments from a historical perspective, we may miss this crucial point (Council of the EU, 1995).
A third characteristic of the literature about privacy is that it is often non-comparative. I mean this in two different senses. On the one hand, there has been very little systematic comparison of how different states with different cultures and institutions have responded to these same technological challenges. We should continually reflect on how the same problems are resolved in different countries, and moreover what this then says about their approach to privacy protection, their institutional capacities to manage technological change, and their national cultures. Data protection is a marvellous issue for these kinds of comparative studies. But so is computer matching, Caller ID, videosurveillance, encryption, genetic privacy and so on. There is enormous scope for analysts to take a technology that has diffused around the advanced industrial world and to investigate and attempt to explain similarities and difference in national responses.
Our discourse is non-comparative in a second sense in that it tends to be discussed as a discrete, compartmentalized issue. It has become what policy analysts call a distinct policy sector characterized by: a separate set of statutory instruments; regulatory bodies established to implement these laws; a circle of legal experts; a small group of journalists ready to publicize information abuses; a growing academic community with expertise in data protection and privacy; and a range of international arenas in which the policy community can exchange ideas and collaborate.
The result of this "sectorisation" of the policy is that very little comparison is made with other related fields. This sectorisation is in the face of some compelling arguments that privacy protection ought to be regarded as one component of a more coherent information policy. Furthermore we often lose sight of the fact that much can be learnt about the implementation of privacy regulations from the analysis of other regulatory fields that have less obvious connections. There are some really interesting cross-sectoral parallels with policies of environmental protection, for example.
Finally, the discourse about the politics of privacy tends to be overly descriptive. We do spend a lot of time and effort describing for each other the nature of our respective privacy laws. We describe their scope and coverage, their treatment of tricky definitional problems, the powers they grant to the commissioners and so on. Comparative legalistic analysis has reached the point where journals such as Privacy Laws and Business and Transnational Data and Communications Report can publish regular and very useful league tables on the worldwide state of data protection law. This approach reflects the predominance of legal scholarship, in which questions are confined to reviewing the state of the law in different countries and pointing out inconsistencies or inadequacies as new technologies enter society and challenge existing regulations.
I do not criticize this. I do think that the emphasis on comparative legalistic analysis carries some dangers. In particular, when we focus too much on the state of the law, we may ignore the fact that the statutory or legal context is just one factor that influences the behaviour of those in public and private organizations that process personal data. There are other more elusive, but sometimes more important, cultural and institutional factors that are seldom recognized.
There are, for example, different administrative cultures in Western societies. Some of these, like those of Germany, are traditionally deferential to rules and administrative norms. I would argue, as does Flaherty, that this administrative culture makes the task of the data protector somewhat easier. In countries such as the United States, on the other hand, there is less of a centralized civil service mentality. Administration is more open to outside interests and personnel. Data protection policy (in the form of the 1974 Privacy Act) has to be implemented within a more fragmented, less hierarchical, administrative system (Gellman, 1993).
Another aspect of a nation's culture can have an important impact on privacy protection. Political scientists have long tried to classify cultures according to whether they are deferential to authority, or more prone to participation, complaint, protest and so on. It is at least hypothetically true that countries whose citizens are less deferential to authority are going to be less trusting of agencies to handle personal information with care and propriety, and more prone to complain to relevant authorities. The cultural context within which data protection law needs to be implemented can, nevertheless, be lost when we focus too narrowly on the wording of law.
Another significant factor overshadowed by the legalistic emphasis of our conversations is the overall structural or institutional arrangements of government. This affects the position of data protection on the political and administrative agenda in a variety of ways. For instance, the relations between political and administrative agencies creates distinctive styles of governance. The more consensual and accommodating styles of countries like Sweden are more conducive to anticipating developments and providing appropriate responses, than the more conflictual, fragmented and unpredictable style of a place like the United States.
These and other institutional and cultural factors have profound comparative influences on the development and implementation of privacy and data protection policy. Sometimes they tend to be forgotten in the discussions over the most appropriate legal and regulatory responses to the latest technological intrusion.
The potential contribution of political science to the study of privacy thus lies in four related areas. First, a range of studies could be conducted on the relationship between political and administrative culture and attitudes toward privacy at both elite and public levels. These different national orientations are in turn hypothetically related to policy responses in different sectors.
Second, we need to come to grips with the question of policy success. How do we know whether public policies on privacy are "working"? What measures of policy success can be empirically determined and presented? What do these measures "measure"? Lessons can be drawn from related regulatory fields to provide insights into these questions.
Third, comparative studies are necessary. These could be cross-national, in which the capacities of different states to manage the same technological challenge are compared and evaluated. They could be cross-technology, in which the responses of one state to different privacy invasive practices are compared (see, Regan, 1995, forthcoming). They could also be cross-sectoral, in which the responses of different economic interests are compared (Bennett, 1995). Comparative analysis is necessary both to escape American ethnocentrism, and to provide insights into larger theoretical questions about the capacity of democratic states to manage technological change.
Studies on any of these questions would help bridge the gap between the conceptual/philosophical and the descriptive/empirical. While a large and diverse range of literature has been incorporated into this review, very few works have seriously tried to advance our understanding through the empirical examination of hypotheses informed by wider theoretical propositions. Thus at the end of this review, I would emphasize what could be written, rather than what has been written. The potential for a political economy of privacy is enormous.
The literature cited in this survey has been compiled from a variety of different sources. A large portion (especially that on data protection) has been cited in previously published works (especially, Bennett, 1992). This has been supplemented by a review of relevant works gleaned through the bibliographic search conducted for the Human Genome Project at the Center for Social and Legal Research, and by suggestions from scholars associated with this project.
A more systematic, though far less useful, search was conducted within the more general political science literature. In particular, the reference publication ABC Pol Sci was consulted for the last ten years and relevant articles noted. Very little, however, was found when the following standard political science journals were searched: The American Political Science Review, the Journal of Politics, the American Journal of Political Science, Comparative Politics, Comparative Political Studies, British Journal of Political Science, Political Science Quarterly, European Journal of Political Research, Canadian Journal of Political Science. A browsing through the indexes of a collection of contemporary introductory textbooks on American politics also revealed a paucity of references to privacy.
ABC Pol Sci: A Bibliography of Contents: Political Science and Government. Santa Barbara, Ca: ABC-CLIO.
Agre, Philip E. 1994. "Surveillance and Capture: Two Models of Privacy," The Information Society 10: 101-127.
Aldrich, Robert. 1982. Privacy Protection Law in the United States. Washington D.C.: U.S. Department of Commerce, National Telecommunications and Information Administration.
Allen, Anita L. 1985. Uneasy Access: Privacy for Women in a Free Society. Totawa, N.J.: Rowman and Littlefield.
Allen, Anita L. & Erin Mack. 1990. "How Privacy Got its Gender," Northern Illinois University Law Review 10: 441-78.
Almond, Gabriel A. 1988. "Separate Tables: Schools and Sects in Political Science," PS: Political Science and Politics 21: 828-42.
Almond, Gabriel A. & Sidney Verba. 1965. The Civic Culture. Boston: Little Brown.
________. 1980. The Civic Culture Revisited. Boston: Little Brown.
Arndt, H.W. 1949. "The Cult of Privacy," Australian Quarterly XXI: 69-71.
Barlow, John Perry. April 1994. "Jackboots on the Infobahn," Wired: 40-8.
Beniger, James R. 1986. The Control Revolution: Technological and Economic Origins of the Information Society. Cambridge: Harvard University Press.
Benn, Stanley I. & Gerald F. Gaus. 1983. eds. Public and Private in Social Life. New York: St. Martin's Press.
Bennett, Colin. J. 1988a. "Different Processes, One Result: The Convergence of Data Protection Policy in Europe and the United States," Governance 1: 415-41.
________. 1988b. "Regulating the Computer: Comparing Policy Instruments in Europe and the United States," European Journal of Political Research 16: 437-66.
________. 1991. "Computers, Personal Data and Theories of Technology: Comparative Approaches to Privacy Protection in the 1990s," Science, Technology and Human Values 16: 51-69.
________. 1992. Regulating Privacy: Data Protection and Public Policy in Europe and the United States. Ithaca: Cornell University Press.
________. 1995. Implementing Privacy Codes of Practice: A Report to the Canadian Standards Association. Rexdale, Ont.: Canadian Standards Association.
________. 1995 forthcoming. "The Public Surveillance of Personal Data," in David Lyon And Elia Zureik eds. New Technology, Surveillance and Social Control. Minneapolis: University of Minnesota Press.
Bensman, Joseph & Robert Lilienfeld. 1979. Between Public and Private: The Lost Boundaries of the Self. New York: Free Press.
Bing, Jon. 1978. "A Comparative Outline of Privacy Legislation," Comparative Law Yearbook 2: 149-81.
Bloustein, Edward J. 1978. "Privacy is Dear at any Price: A Response to Professor Posner's Economic Theory? Georgia Law Review 12: 429-
Brenton, Myron. 1964. The Privacy Invaders. New York: Coward-McCann.
Bull, Hans Peter. 1984. Datenschutz oder die Angst vor Dem Computer. Munchen: Piper.
Burkert, Herbert. 1982. "Institutions of Data Protection: An Attempt at a Functional Explanation of European National Data Protection Laws," Computer Law Journal 3: 167-88.
Burnham, David. 1980. The Rise of the Computer State. New York: Random House.
Campbell, Duncan and Steve Connor. 1986. On the Record: Surveillance, Computers and Privacy. London: Michael Joseph.
Canada, Privacy Commissioner. 1992. Genetic Testing and Privacy. Ottawa: Minister of Supply and Services.
________. 1993. Annual Report 1992-93. Ottawa: Canada Communication.
Chaum, David. 1985. "Security without Identification: Transaction Systems to Make Big Brother Obsolete," Communications of the ACM 28: 1030-44.
________. 1992. "Achieving Electronic Privacy," Scientific American August 1992: 96-101.
Clarke, Roger A. 1989. "Information Technology and Dataveillance," Communications of the ACM 31: 498-512.
Council of the European Union (EU). 1995. Common Position on the Protection of Individuals in Relation to the Processing of Personal Data and the Free Movement of Such Data. Brussels: European Union.
Council of Europe (CoE). 1981. Convention on the Protection of Individuals with Regard to Automatic Processing of Personal Data. Strasbourg: Council of Europe.
Crick, Bernard. 1964. In Defence of Politics.
London: Penguin Books.
Culnan, Mary J. "How Did They Get My Name? An Exploratory Study of Consumer Attitudes Toward Secondary Information Use," MIS Quarterly 17: 341-63.
Dandeker, Christopher. 1990. Surveillance, Power and Modernity: Bureaucracy and Discipline from 1700 to the Present Day. New York: St.Martin's Press.
Danziger, James N., William H. Dutton, Rob Kling, and Kenneth L. Kraemer. 1982. Computers and Politics: High Technology in American Local Governments. New York: Columbia University Press.
Data Protection Registrar (UK). 1993. Ninth Report. London: HMSO.
Davies, Simon. 1992. Big Brother: Australia's Growing Web of Surveillance. Sydney: Simon & Schuster.
Dutton, William & Robert Meadow. 1985. "Public Perspectives on Government Information Technology: A Review of Survey Research on Privacy, Civil Liberties and the Democratic Process." Report prepared for the US Congress Office of Technology Assessment (OTA). Washington DC: OTA.
Easton, David. 1965. A Framework for Political Analysis. Englewood Cliffs: Prentice-Hall.
Eaton, Joseph W. 1986. Card-Carrying Americans: Privacy, Security and the National ID Card Debate. Totowa, NJ: Rowman & Littlefield.
Ellul, Jacques. 1964. The Technological Society. New York: Vintage Books.
Ervin, Senator Samuel J. Jr. 1971. "Privacy and Government Investigations," University of Illinois Law Forum: 138-150.
Etzioni, Amatai. 1994. Spirit of Community: Rights, Responsibilities and the Communitarian Agenda. New York: Simon and Schuster.
Flaherty, David H. 1967. Privacy in Colonial New England. Charlottesville: University Press of Virginia.
________. 1979. Privacy and Government Data Banks: An International Perspective. London: Mansell.
________. 1984a. Nineteen Eighty-Four and After. Final Report of the Bellagio Conference on Current and Future Problems of Data Protection. London: University of Western Ontario Privacy Project.
________. 1984b. "The Need for an American Privacy Protection Commission," Government Information Quarterly 1: 235-58.
________. 1989. Protecting Privacy in Surveillance Societies: The Federal Republic of Germany, Sweden, France, Canada, and the United States. Chapel Hill, N.C.: University of North Carolina Press.
Foucault, Michel. 1979. Discipline and Punish: The Birth of the Prison. New York: Vintage Books.
Freedman, Warren. 1987. The Right of Privacy in the
Computer Age. New York: Quorum Books
Fried, Charles. 1968. "Privacy," Yale Law Journal 77: 475-93.
Gallati, Robert R.J. 1971. "The New York State Identification and Intelligence System," in Alan F. Westin, Information Technology in a Democracy. Cambridge: Harvard University Press.
Gandy, Oscar. 1992. The Panoptic Sort. Boulder: Westview Press.
Gavison, Ruth. 1980. "Privacy and the Limits of the Law." Yale Law Journal 89: 421-71.
Gellman, Robert M. 1993. "Fragmented, Incomplete and Discontinuous: The Failure of Federal Privacy Regulatory Proposals and Institutions," Software Law Journal VI: 199-238.
Giddens, Anthony. 1985. The Nation State and Violence. Berkeley: University of California Press.
Gould, J.P. 1980. "Privacy and the Economics of Information," Journal of Legal Studies 9: 827
Harris, Louis, & Alan F. Westin. 1979. The Dimensions of Privacy: A National Opinion Research Survey of Attitudes toward Privacy. Stevens Point, Wisc: Sentry Insurance.
________. 1990. The Equifax Report on Consumers in the Information Age. Atlanta: Equifax Inc.
________. 1992a. Harris-Equifax Consumer Privacy Survey. Atlanta: Equifax Inc.
________. 1992b. The Equifax Canada Report on Consumers and Privacy in the Information Age. Montreal: Equifax Canada Inc.
________. 1994. The Equifax Canada Report on Consumers and Privacy in the Information Age. Montreal: Equifax Canada Inc.
Hewitt, Patricia. 1977. Privacy: The Information Gatherers. London: National Council for Civil Liberties.
Hixson, Richard F. 1987. Privacy in a Public Society:
Human Rights in Conflict. New York: Oxford University Press.
Hondius, Frits W. 1975. Emerging Data Protection in Europe. Amsterdam: North Holland.
Hood, Christopher C. 1986. The Tools of Government. Chatham, N.J.: Chatham House.
Hunter, Lawrence & James Rule. 1994. "Toward Property Rights in Personal Information." Unpublished paper, January 15, 1994.
Inness, Julie C. 1992. Privacy, Intimacy, and
Isolation. New York: Oxford University Press.
Katz, James E. 1987. "Telecommunications and Computers: Whither Privacy Policy?" Society 25: 81-6.
Kingdon, John W. 1984. Agendas, Alternatives, and Public Policies. Boston: Little Brown.
Lasswell, Harold. 1950. Politics: Who Gets What, When, How. New Haven: Yale University Press.
Laudon, Kenneth C. 1986. Dossier Society. New York: Columbia University Press.
________. 1994. "Markets and Privacy."
NYU Center for Research on Information Systems, January 1994.
Lawson, Ian. 1995. Privacy and the Information Highway: Regulatory Options for Canada. Ottawa: Industry Canada.
Lemond, Alan and Ron Fry. 1975. No Place to Hide: A Guide to Bugs, Wiretaps, Surveillance, and other Privacy Invasions. New York: St. Martin's.
Linowes, David F. 1989. Privacy in America: Is Your Private Life in the Public Eye? Urbana: University of Illinois Press.
Linowes, David F., and Colin J. Bennett. 1986. "Privacy: Its Role in Federal Government Information Policy," Library Trends 31: 19-42.
Lipset, Seymour Martin. 1990. Continental Divide: The Values and Institutions of the United States and Canada. New York: Routledge.
Long, Edward V. 1967. The Intruders: The Invasion of Privacy by Government and Industry. New York: Praeger.
Lyon, David. 1992. "The New Surveillance: Electronic Technologies and the Maximum Security Society," Crime, Law and Social Change 18: 159-172.
________. 1994. The Electronic Eye: The Rise of Surveillance Society. Minneapolis: University of Minnesota Press.
Lyon, David & Elia Zureik. 1995 forthcoming. New Technology, Surveillance and Social Control. Minneapolis: University of Minnesota Press.
Madgwick, Donald. 1968. Privacy Under Attack. London: National Council for Civil Liberties.
Madgwick, Donald & Tony Smythe. 1974. The Invasion of Privacy. London: Pitman.
Marchand, Donald A. 1980. The Politics of Privacy, Computers and Criminal Justice Records: Controlling the Social Costs of Technological Change. Arlington Va: Information Resources Press.
Marx, Gary. 1988. Undercover: Police Surveillance in America. Berkeley: University of California Press.
Marx, Gary & Nancy Reichman. 1984. "Routinizing the Discovery of Secrets: Computers as Informants," American Behavioral Scientist 27: 423-452.
Mazmanian, Daniel A. & Paul A. Sabatier. 1983. Implementation and Public Policy. Glenview, Ill: Scott, Foresman.
McCloskey, Herbert J. 1980. "Privacy and the Right to Privacy," Philosophy 55: 17-38.
Miller, Arthur R. 1971. The Assault on Privacy: Computers, Data Banks and Dossiers. Ann Arbor: University of Michigan Press.
Moore, Barrington Jr. 1984. Privacy: Studies in Social and Cultural History. Armonk, NY: M.E. Sharpe.
Neier, Aryeh. 1975. Dossier: The Secret Files They Keep on You. New York: Stein and Day.
Noam Eli M. 1994. Privacy in Telecommunications: Markets, Rights and Regulations. Monograph in "Ethics in Telecommunication" Series. Cleveland: Office of Communication, United Church of Christ.
O'Brien, David M. 1979. Privacy, Law, and Public Policy. New York: Praeger.
Organization for Economic Cooperation and Development. 1981. Guidelines on the Protection of Privacy and Transborder Flows of Personal Data. Paris: OECD.
Packard, Vance. 1964. The Naked Society. New York: David McKay.
Parent, William A. 1983. "Privacy, Morality and the Law," Philosophy and Public Affairs 12: 269-88.
Parker, Richard. 1974. "A Definition of Privacy," Rutgers Law Review 27: 275-96.
Pateman, Carole. 1970. Participation and Democratic Theory. Cambridge: Cambridge University Press.
_______. 1983. "Feminist Critiques of the Public/Private Dichotomy," in S. I. Benn & G.F. Gaus eds. Public and Private in Social Life. New York: St.Martin's Press.
Pennock, J. Roland & John W. Chapman. 1971. Privacy. Nomos XIII. New York: Atherton Press.
Plesser, Ronald L. 1991. Privacy Protection in the United States. Washington D.C.: Piper & Marbury.
Pool, Ithiel de Sola. 1983. Technologies of Freedom: On Free Speech in the Electronic Age. Cambridge: Harvard University Press.
Posner, Richard A. 1978a. "An Economic Theory of Privacy," Regulation May-June 1978: 19-26.
________. 1978b. "The Right to Privacy," Georgia Law Review 12: 393-422.
Poster, Mark. 1990. The Mode of Information. New York: Polity Press.
Powers, Madison. 1993. "The Right to Privacy Reconsidered." (Unpublished paper)
________. 1994. "Privacy and the Control of Genetic Information," in Mark S. Frankel & Albert Teich, The Genetic Frontier: Ethics, Law and Policy. Washington DC: American Association for the Advancement of Science.
Prosser, William. 1960. "Privacy," California Law Review 48: 383-423.
Raab, Charles. 1993. "Data Protection in Britain: Governance and Learning," Governance 6: 43-66.
Raab, Charles D. & Colin J. Bennett. 1993. "Measuring the Performance of Data Protection Systems." Paper presented to the 6th Annual Conference of Privacy, Laws and Business, St. Catherine's College, Oxford.
________. 1994. "Protecting Privacy across Borders: European Policies and Prospects." Public Administration 72: 95-112.
Raines, John Curtis. 1974. Attack on Privacy. Valley Forge, PA: Judson Press.
Regan, Priscilla M. 1981. "Public Uses of Private Information: A Comparison of Personal Information Policies in the United States and Britain." Ph.D., Cornell University.
________. 1984. "Personal Information Policies in the United States and Britain: The Dilemma of Implementation Considerations," Journal of Public Policy 4: 19-38.
________. 1990. "Protecting Privacy and Controlling Bureaucracies: Constraints of British Constitutional Principles." Governance 3: 33-54.
________. 1993a. "Surveillance and New Technologies: Changing Nature of Workplace Surveillance," Paper presented to the Strategic Research Workshop on "New Technology, Surveillance and Social Control," Queen's University, May 1993.
________. 1993b. "Ideas of Interests: Privacy in Electronic Communications," Policy Studies Journal 21: 450-469.
________. 1994. "The Globalization of Privacy: Implications of Recent Changes in Europe." American Journal of Economics and Sociology 52: 257-74.
________. 1995 (forthcoming). Privacy, Technology and Public Policy. Chapel Hill: University of North Carolina Press.
Rosenberg, Jerry M. 1969. The Death of Privacy. New York: Random House.
Rotenberg, Marc. 1991. Privacy Law in the United States: Failing to Make the Grade. Washington D.C.: Computer Professionals for Social Responsibility.
________. 1994. "Privacy: Political Right or Economic Good," Address given to the 1994 Conference on Computers, Freedom and Privacy (CFP '94), Chicago, March 1994.
Rothfeder, Jeff. 1992. Privacy for Sale. New York: Simon and Shuster.
Rubin, Michael Rogers. 1988. Private Rights, Public Wrongs: The Computer and Personal Privacy. New Jersey: Ablex.
Rule, James B. 1974. Private Lives and Public Surveillance: Social Control in the Computer Age. New York: Schocken Books.
Rule, James, Douglas MacAdam, Linda Stearns, and David Uglow. 1980. The Politics of Privacy: Planning for Personal Data Systems as Powerful Technologies. New York: Elsevier.
Samar, Vincent J. 1991. The Right to Privacy: Gays, Lesbians and the Constitution. Philadelphia: Temple University Press.
Schoeman, Ferdinand D. ed. 1984. Philosophical Dimensions of Privacy: An Anthology. Cambridge: Cambridge University Press.
Schwartz, Paul. 1989. "The Computer in German and American Constitutional Law: Towards an American Right of Information Self-Determination," American Journal of Comparative Law 37: 675-701.
________. 1992. "Data Processing and Government Administration: The Failure of the American Legal Response to the Computer," Hastings Law Journal 43: 1321-88.
Seipp, David J. 1978. The Right to Privacy in American History. Cambridge Mass.: Harvard University Program on Information Resources Policy.
Shils, Edward. 1956. The Torment of Secrecy. Glencoe, Ill: Free Press.
Sieghart, Paul. 1976. Privacy and Computers. London: Latimer.
Simitis, Spiros. 1987. "Reviewing Privacy in an Information Society," University of Pennsylvania Law Review 135: 707-46.
Simitis, Spiros, Ulrich Dammann, Otto Mallmann and Hans Reh. 1981. Kommentar zum Bundesdatenschutzgesetz. 3d ed. Baden-Baden: Nomos.
Smith, H. Jeff. 1994. Managing Privacy: Information Technology and Corporate America. Chapel Hill: University of North Carolina Press.
Smith, Robert Ellis. 1979. Privacy: How to Protect What's Left of It. Garden City, NJ: Anchor Press.
________. 1992. Compilation of State and Federal Privacy Laws. Providence RI: Privacy Journal.
________. 1993. War Stories: Accounts of Persons Victimized by Invasions of Privacy. Providence RI: Privacy Journal.
Sobel, Lester A. ed. War on Privacy. 1976. New York: Facts on File Inc.
Stigler, George J. 1980. "An Introduction to Privacy in Economics and Politics," Journal of Legal Studies 9: 623
Stone, Michael & Malcolm Warner. 1969. "Politics, Privacy and Computers," The Political Quarterly 40: 260
Stouffer, Samuel. 1955. Communism, Conformity and Civil Liberties. New York: Doubleday.
Sullivan, John L., Piereson, James E., & Marcus, George E. 1982. Political Tolerance and American Democracy. Chicago: University of Chicago Press.
Thompson, Anthony A. 1970. Big Brother in Britain Today. London: Michael Joseph.
Tilly, Charles ed. 1975. The Formation of National States in Western Europe. Princeton: Princeton University Press.
U.S. Department of Health, Education and Welfare (HEW). Records, Computers and the Rights of Citizens. Report of the Secretary's Advisory Committee on Automated Personal Data Systems. Washington D.C.: HEW, 1973.
U.S. Congress. Office of Technology Assessment (OTA), 1982. Computerized Criminal History Systems. Washington D.C.: US Government Printing Office.
________. 1986. Electronic Record Systems and Individual Privacy. Washington D.C.: US Government Printing Office.
U.S. Privacy Protection Study Commission (PPSC). 1977. Personal Privacy in an Information Society. Washington, D.C.: Government Printing Office.
Warner, Malcolm and Michael Stone. 1970. The Data Bank Society: Organizations, Computers, and Social Freedom. London: Allen & Unwin.
Warren, Samuel and Louis Brandeis. 1890. "The Right to Privacy," Harvard Law Review 4: 193-220.
Westin, Alan F. 1967. Privacy and Freedom. New York: Atheneum.
________. 1980. "Introduction," In Donald A. Marchand, The Politics of Privacy, Computers and Criminal Justice Records, Arlington Va: Information Resources Press.
________. 1994. "Privacy and Genetic Information: A Sociopolitical Analysis," in Mark S. Frankel & Albert Teich eds. The Genetic Frontier: Ethics, Law and Policy. Washington DC: American Association for the Advancement of Science.
Westin, Alan F. and Michael A. Baker. 1972. Databanks in a Free Society: Computers, Record-Keeping, and Privacy. New York: Quadrangle Books.
Will, Ian. 1983. The Big Brother Society. London: Harrap.
Winner, Langdon. 1977. Autonomous Technology: Technics-out-of-Control as a Theme in Political Thought. Cambridge: MIT Press.