The Problem of Privacy in Public September 28, 2000 1
Protecting Privacy in an Information Age:
The Problem of Privacy in Publici
Helen Nissenbaum
University Center for Human Values
5 Ivy Lane
Princeton University
Princeton, NJ 08544
Law and Philosophy, 17: 559-596, 1998
ABSTRACT. Philosophical and legal theories of privacy have long recognized the relationship
between privacy and information about persons. They have, however, focused on personal, intimate,
and sensitive information, assuming that with public information, and information drawn from public
spheres, either privacy norms do not apply, or applying privacy norms is so burdensome as to be
morally and legally unjustifiable. Against this preponderant view, I argue that information and
communications technology, by facilitating surveillance, by vastly enhancing the collection, storage,
and analysis of information, by enabling profiling, data mining and aggregation, has significantly
altered the meaning of public information. As a result, a satisfactory legal and philosophical
understanding of a right to privacy, capable of protecting the important values at stake in protecting
privacy, must incorporate, in addition to traditional aspects of privacy, a degree of protection for
privacy in public.
KEY WORDS: privacy, search and seizure, balancing rights, information technology
INTRODUCTION
There is growing awareness as well as resentment of the routine practice of recording,
analyzing, and communicating information about individuals as they act and transact in the
normal course of their commercial and public lives. The information in question is taken into
the possession of and used by whomever collects it and from there may be transmitted—
usually electronically, usually for fee or favor—to others—second parties, third parties,
fourth parties, and so on. While philosophical theories have long acknowledged the
relationship between privacy and information about persons, and have argued for limits on
allowable practices of information gathering, analyzing, and sharing as a means of protecting
privacy, their efforts have primarily applied to intimate and sensitive information.
The Problem of Privacy in Public September 28, 2000 2
While not denying the importance of protecting intimate and sensitive information,
this paper insists that theories of privacy should also recognize the systematic relationship
between privacy and information that is neither intimate nor sensitive and is drawn from
public spheres. The significance of this information for privacy has emerged in recent
decades as a result of contemporary surveillance practices enabled by advances in
information technology, creating what I here call the problem of privacy in public.ii
As
observed in 1985 by Larry Hunter, a computer scientist, “Our revolution will not be in
gathering data—don’t look for TV cameras in your bedroom—but in analyzing the
information that is already willingly shared.”iii
In the course of this paper I will argue that privacy in public, which in the past has
been explicitly excluded or merely neglected by many of the most highly-regarded and oftencited
philosophical and legal works on privacy, is a genuine privacy interest that is worthy of
study as well as protection.
The discussion proceeds as follows. After surveying circumstances and activities that
give rise to the problem of privacy in public, I offer an explanation for why predominant and
influential theoretical accounts of privacy have failed to deal explicitly with it. Following
this, in what may be seen as the core of the paper, I identify the features of contemporary
surveillance practices that are central to viewing these practices as genuine concerns for any
normative theory of privacy. In the concluding sections of the paper, I consider how we may
absorb privacy in public into comprehensive theories of privacy. Although I do not provide
such a theory myself, I suggest that resources are already present in some existing theories –
for example, in work by Ferdinand Schoeman and, more recently, by Judith DeCewiv
. I also
clear the way for such a theory by showing how certain barriers that, in the past, have
seemed insurmountable may be overcome.
I. THE PROBLEM OF PRIVACY IN PUBLIC
My interest in the problem of protecting privacy in public is motivated by circumstances in
the real world that are obviously problematic for most people, and have frequently been
reported in public and popular mass media.v
These circumstances are that even, and
The Problem of Privacy in Public September 28, 2000 3
especially, in the public arena, people have become targets of surveillance at just about every
turn of their lives. In transactions with retailers, mail order companies, medical care givers,
daycare providers, and even beauty parlors, information about them is collected, stored,
analyzed and sometimes shared. Their presence on the planet, their notable features and all
their momentous milestones are dutifully recorded by agencies of federal, state and local
government including birth, marriage, divorce, property ownership, drivers’ licenses, vehicle
registration, moving violations, parenthood, and, finally, their demise. Into the great store of
information, people are identified through name, address, phone number, credit card
numbers, social security number, passport number, and more; they are described by age, hair
color, eye color, height, quality of vision, mail orders and on site purchases, credit card
activity, travel, employment history, rental history, real estate transactions, change of
addressvi
, ages and numbers of children, and magazine subscriptions.vii
The dimensions are
endless.
In several ways, information technology is essentially implicated in this relentless
gathering of information. In the first place, computerized databases have provided for it the
right kind of home. Information that is drawn from the physical world is harbored in
electronic databases, which give these records the permanence, malleability and
transportability that has become the trademark of information technology. Without
information technology, the gatherers and users of information would be able neither to
conduct surveillance (that is, gather the data), nor create databases of great magnitude and
power, nor extract the information that motivates these activities. Roughly forty years ago,
this application of information technology to the creation of computerized databases mainly
by government and other large organizations, was the first to attract concern among policy
analysts, journalists and fiction writers.
In the unfolding of recent developments in information technology, and especially
comprehensive digital electronic networks, there is another means by which information may
be harvested. In contemporary, technologically advanced societies, it is commonplace for
large sectors of populations to participate, in varying degrees, in electronically networked
interactions. Governments, as well as individual and institutional agents of the private
sector, encourage such participation by their explicit expressions of approval, by
The Problem of Privacy in Public September 28, 2000 4
progressively increasing the ease of access, as well as speed and declining prices (for
example, through the World Wide Web), and at the same time creating the possibility for
more and more to be done by electronic means. Once in the electronic sphere, the tracks of
people's activities may be recorded directly into electronic databases. Electronic
transactions, even carefree meanderings (popularly referred to as "browsing" and "surfing")
may be captured and recorded.viii
Information like email addresses, system characteristics,
and a trail of network-based activities are not only effortlessly recorded, but easily combined
with information from the physical world. In this activity information technology is doubly
implicated as it acts as the medium for transactions as well as repository for the
information.ix
In addition to these two means by which information technology facilitates
surveillance, there is yet another layer of surveillance that builds upon them. Where most of
the activities earlier described involved the collecting of information by an agency,
organization, or individual with whom a person interacts directly, this new layer involves
secondary users and suppliers who acquire information from other sources, either the
primary sources or other secondary sources. These secondary, or second-order purveyors of
information include credit bureaus—and the so-called "super-bureaus”—medical insurance
bureaus, and list brokers.x
Although some of the information supplied to agents of
secondary collection is drawn from the private sector, including banks, credit card
companies, and retailers, much is drawn from government records. No longer is it necessary
to send a person to a court house to copy these records, painstakingly, into databases.
The electronic format offers great convenience and flexibility; databases may be
searched for individual records or entire databases may be transferred via digital electronic
networks. Some government agencies are fast understanding that their computerized
records may be a source of significant revenue.xi
But even when they have balked at the idea
of releasing information electronically, courts have forced them to do so.xii
Secondary
harvesting of information is held deeply under suspicion not only because it is seen as the
significant driver of the unquenchable thirst for information about persons as well as its
seemingly endless supply, but also because people perceive it to be illegitimate.
The Problem of Privacy in Public September 28, 2000 5
This uncontrolled harvesting of public information has not escaped the notice of
scholars and advocates of policy, who consider it a serious problem for privacy that public
as well as corporate policy has not adequately addressed. Although the privacy concerns of
data subjects have not been completely ignored in the policy arena, they are more often
noticed as a result of a highly publicized media event than as a result of thoughtful public
deliberation over the need for privacy. A case in point is the Video Privacy Protection Act
(known commonly as the "Bork Bill"). When a national newspaper published the video
rental records of Robert Bork during Senate Hearings for his nomination as Associate
Justice for the Supreme Court, Congress hastily responded with the Video Privacy Act. xiii
The result is a body of policy that is piecemeal and inconsistent.xiv
As disturbing as the practices of public surveillance are, they seem to fall outside the
scope of predominant theoretical approaches to privacy, which have concerned themselves
primarily with two aspects of privacy—namely, maintaining privacy against intrusion into
the intimate, private realms, and protecting the privacy of individuals against intrusion by
agents of government. Philosophical and legal theories of privacy offer little by way of an
explicit justificatory framework for dealing with the problem of privacy in public. Indeed,
with only a few exceptions, work within these traditions appears to suffer a theoretical blind
spot when it comes to privacy in public, for while it has successfully advanced our
understanding of the moral basis for privacy from some of the traditionally conceived
threats, such as violation of the personal sphere, abuse of intimate information, protection of
the private individual against government intrusion, and protection of, say doctor-patient,
lawyer-client and similar special relationships, it has not kept abreast of the privacy issues
that have developed in the wake of advanced uses of information technology.
Although Hunter, in the passage quoted earlier, may have understated the extent that
the sheer growth in data gathering affects privacy and the extent to which technological
means allows intrusion into and surveillance of even private, enclosed spaces,xv
he accurately
predicted not only that analysis of information will be a major source of privacy invasion,
but that because the information analyzed is willingly shared, people are, in some sense,
complicit in the violation of their own privacy. Accordingly, although the traditional topics
covered by philosophical discussions remain important both for their historical significance
The Problem of Privacy in Public September 28, 2000 6
and their present urgency and seriousness, they no longer cover the full extent of a need for
privacy protection in our information age where the practice of public surveillance, record
keeping, and information analysis seems to be growing not only without apparent limit but
so completely out of the control of those who are its subjects.
This paper’s emphasis on theoretical and conceptual foundations of privacy—not
public or business policy—does not preclude consideration of important practical
implications. In particular, I would suggest that the absence of a clearly articulated
philosophical base is not of theoretical interest only, but is at least partially responsible for
the inconsistencies, discontinuities and fragmentation, and incompleteness in the framework
of legal protections and in public and corporate policy. It may be useful to consider the
practical import of an inadequately developed conceptual scheme in terms of an actual
case—the case of Lotus Marketplace.
In April 1990, Lotus Development Corporation, a developer and marketer of
popular software, and Equifax Inc., one of the “big three” companies that collect and sell
information about consumer financial transactions,xvi
announced their intention to produce a
database called "Lotus Marketplace: Households" which would contain actual and inferred
information about approximately 120 million individuals in the United States. It would
include name, address, type of dwelling, marital status, gender, age, household income,
lifestyle, and purchasing propensity. The two companies expected the database, which was
to have been recorded and sold in the format of a CD-ROM, to be widely adopted by
marketers and mailing companies.xvii
Grassroots opposition, including an estimated 30,000
letters of protest, led company executives to announce, in January 1991, that they were
canceling the project. Even as privacy advocates and individual participants trumpeted
victory for privacy, executives insisted that their actions were prompted only by negative
publicity and public misunderstanding and not by a conviction of wrongdoing. They insisted
that their product would not have violated privacy.
Though hailed as a victory for privacy, the legacy of Lotus Marketplace Households
for the course of data gathering has been negligible; current practices far surpass it in scope
and magnitude. This result suggests that in the absence of well understood and clearly
articulated normative principles, the decision to withdraw Lotus Marketplace Households,
The Problem of Privacy in Public September 28, 2000 7
by itself, provides a scant basis for dealing with subsequent challenges.xviii
There was no
common agreement that here was an effort that violated privacy, or an understanding of the
reasons why it violated privacy. The same may be said for the other individual victories that
the dogged efforts of policy advocates have yielded. With no underlying thread to tie one
effort to another, each must be fought on its own terms; the fate of privacy in public remains
in the hands of those with the most energy and with the strongest lobbies; it does not reflect
underlying values at all.
II. WHY PRIVACY IN PUBLIC IS DISMISSED
Before responding directly to the challenge of producing principles by which Lotus
Marketplace Households and similar efforts may be judged violations of privacy, I consider
the reasons why many influential philosophical theories of privacy may not have addressed
directly the cluster of issues raised by widespread public surveillance. If privacy in public
does constitute a genuine privacy interest, then not only is it important to construct the
much needed justificatory framework, but also to ask why philosophical and normative
theories of privacy have either explicitly dismissed the idea of any genuine privacy interest in
public, or merely have overlooked it.xix
A variety of factors have shaped normative theories of privacy, making them more
responsive to some types of problems and constraints and less responsive to others.
Examining these theories with a view to understanding why specifically they either neglect
or dismiss the normative force of privacy in public, three factors (there may be others)
emerge, which I have labeled, respectively, conceptual, normative, and empirical.
Conceptual
To many, the idea that privacy may be violated in public has an oddly paradoxical
ring. One likely source of this response is the way the terms "public" and "private" have
been used in political and legal theory. Although their respective meanings may vary from
one context to another (and I take it this assertion is relatively uncontroversial among
scholars in these areas), the terms are almost always used as a way to demarcate a strict
The Problem of Privacy in Public September 28, 2000 8
dichotomy of realms.xx
In some contexts, for example, the term "private" indicates the realm
of familial and other personal or intimate relations, while the term "public" indicates the civic
realm or realm of community outside of this personal one. In some contexts, "public"
indicates the realm of governmental institutions in contrast with the realm of "private"
citizens or "private" institutions (such as corporations). In relation to law, the term "private"
generally marks a distinctive area dedicated to settling scores between people in their
capacities as private citizens, in contrast with "public" law, which generally covers disputes
in which officials or agencies of government are involved. In a similar vein Judith W.
DeCew observes,
The public/private distinction has sometimes been taken to reflect differences
between the appropriate scope of government, as opposed to self-regulation by
individuals. It has also been interpreted to differentiate political and domestic
spheres of life. These diverse linguistic descriptions capture overlapping yet
nonequivalent concepts. Nevertheless they share the assumption that there is a
boundary marking off that which is private from that which is public.xxi
For the majority of theorists, it follows seamlessly that the concept and value of privacy
corresponds with, or applies to, the sphere of the private alone. In the past few decades,
therefore, the issues most vigorously pursued in philosophical and legal work on privacy, the
defenses of privacy most thoroughly articulated, are remarkably consonant with these
dichotomies—as I briefly illustrate below.
Following the lines of the private/public dichotomy as it identifies distinctive realms
of individual citizens and private sector institutions versus governmental agents and
institutions, there is a substantial body of work by philosophers, as well as legal and political
theorists, scholars and advocates of policy, and novelists, who have viewed privacy as an
effective way to keep government out of the lives of private individuals and institutions.
Historically, this impulse has made perfect sense in light of government’s enthusiasm for
using computerized databases as a means of storing records of information about people.
Certainly government had the resources and manpower as well as the need to apply the
power of computing to the substantial corpus of personal information that it routinely
The Problem of Privacy in Public September 28, 2000 9
collects.xxii
In 1965, when, in the name of efficiency and efficacy, the Social Science
Research Council, proposed a Federal Data Center to coordinate government statistical
information, critics were immediately alert to the political and personal threat implicit in this
proposal.xxiii
A great deal of the research and scholarship on privacy that immediately followed
this period focused on privacy as a means of maintaining the traditionally valued balance of
power between government and private individuals. This work connects the concept and
value of privacy with the considerable body of theoretical work on the relationship of
individuals in political society to government. It has been able to promote the value of
privacy by showing that privacy is an important means by which individuals may sustain
power, liberty, and autonomy against potentially overwhelming forces of government. Being
able to draw on traditional thinking about the balance of power, has helped advocates and
scholars gain support for public policy to constrain and control government record-keeping
practices. Powerful fictional images such as Big Brother, developed in George Orwell's
novel 1984, together with observed experiences of life under totalitarian regimes, have lent
credence to the practical efforts of privacy advocates.
In parallel with the private-public dichotomy that marks distinct realms of the
intimate or sensitive, on the one hand, and the non-intimate, on the other, there is a
considerable body of work by philosophers and others that argues for protection of intimate
and sensitive realms against intrusion by government or any other individual or collective
agent. This work assumes the existence of distinctive realms of the personal, familial, and
intimate, on the one hand, contrasted with the public, on the other. Scholars interested in
this form of privacy protection emphasize the importance of a realm to which people may
go, from which others are excluded. They conceive of this realm in terms of a secure
physical space, in terms of a private psychological space, or even in terms of a class of
information that is sensitive or intimate over which one would have supreme control.
Those who emphasize the importance of an intimate zone or sphere would say that
defending the integrity of this private realm is a means of enhancing other goods, such as
autonomy, liberty, personal relationships, and trust. Defenders suggest these goods may be
either necessarily or empirically dependent on an individual's having sovereignty over an
The Problem of Privacy in Public September 28, 2000 10
intimate realm.xxiv
Thus, theorists invest privacy with value by showing that privacy
preserves these universally recognized values.
In this section, I have tried to show that the dichotomy between private and public
naturally leads to certain lines of inquiry into privacy. While the dichotomy between public
and private has yielded some important insights into the role and value of privacy, it has
diverted attention from others. It does so by establishing conceptual categories that are not
only hard to bridge but carry with them the implication that privacy is an interest we need
protect in the private realm alone and, by implication, that privacy in public makes little
sense at all. To the extent that a public-private dichotomy drives the direction of theory and
policy, it naturally leads to a concentration on the private sphere alone and—mistakenly, I
think—has made the idea of privacy in public seem paradoxical.
Normative
If conceptions of the public-private dichotomy have implicitly or explicitly affected the
agenda for privacy theory by placing some issues in the limelight and others backstage,
modes of normative argumentation have lent plausibility to certain dimensions of the privacy
interest while seeming to expose others as indefensible. Claims for the protection of privacy
in public have fallen into the second category as they have appeared fatally vulnerable to a
persistent and apparently "knock-down" objection which refers to overriding competing
interests. How so?
It is common for theorists and advocates of privacy to agree that while privacy is an
important interest it must be balanced against other, competing interests. (This strategy is, of
course, not unique to privacy.) While theorists, in their distinctive ways, have argued that
privacy ought to be protected, they have understood that protecting privacy for one person
inevitably leads to restraints on the freedom of another or others, or may even result in
harms to them. Even those generally sympathetic to the idea of a moral right to privacy have
been ready to moderate the exercise of this right in light of some of these competing claims.
Privacy in public is frequently a victim of such balancing as it regularly succumbs to the
apparently overwhelming weight of competing interests.
The Problem of Privacy in Public September 28, 2000 11
A crisp version of this objection may be found in Jeffrey Reiman’s paper, "Privacy,
Intimacy and Personhood,"xxv
Reiman, who characterizes privacy as a social practice
involving "a complex of behaviors that stretches from refraining from asking questions about
what is none of one's business to refraining from looking into open windows one passes on
the street" xxvi
and who argues that privacy is essential for the formation of a conception of
the self, nevertheless concedes that the social practice of privacy "does not assert the right
never to be seen even on a crowded street."xxvii
This concession, in one form or another, is
at bottom of the persistent normative objection that has so effectively blocked attempts to
protect privacy in public.
The power of this widely used rejoinder rests in a foundation of considerations that
have been intuitively compelling to many. One is that claims in favor of privacy in public
affect information that is ostensibly innocuous, namely, information we would not normally
judge to be sensitive or intimate. This being so, it does not take much for a person's claim to
privacy with respect to this information to be outweighed by countervailing claims, even
ones that themselves are not terribly weighty. Another consideration is that if people make
no effort to cover, hide, or remove themselves, or information about themselves, from
public view, if they willingly yield information into the public domain, then they have “let the
cat out of the bag.” It is unreasonable of them to think that, having let the information out,
they can subsequently shift course and “get it” back, suppress it.xxviii
If, for example, you
stroll downtown wearing a red sweater, then you have freely exposed the information that
you were wearing a red sweater at a certain time and date. It is unreasonable to expect that
this information may later be suppressed.
Not only is this unreasonable, but it is wrong because it imposes an unacceptable
restraint on the freedom of others. If you have chosen to expose yourself and information
about yourself in public view with the result that others have access to you, or to
information about you without intruding upon your private realm, then any restrictions on
what they may observe, record and do with this information cannot be justified. In the case
of your red sweater, you could not, for example, expect others to avert their gaze so as not
to see what you were wearing. You could not stop them remembering what you were
wearing, nor prevent them from telling others about it. Such requirements would amount to
The Problem of Privacy in Public September 28, 2000 12
an excessive restraint on the freedoms of others to observe, speak (about your red sweater),
and possibly even profit from so doing. Applying the relevant phrase in legal discourse, a
critic might say that because in a public area we have no “reasonable expectation of
privacy,” we have no right to limit access of others to the information we there expose.
These considerations have held enormous power in theoretical discussions of privacy
and, to my knowledge, have rarely been directly challenged.xxix
In Charles Fried’s influential
paper on privacy, for example, although he defends a robust moral and legal right to privacy,
he is equally explicit about its limits. On the one hand he argues that a right to privacy, a
right to control information about oneself, ought to be secured through law because: "By
using the public, impersonal and ultimate institution of law to grant persons this control, we
at once put the right to control as far beyond question as we can and at the same time show
how seriously we take that right.”xxx
On the other hand, although a right to privacy would
be recognized by law, it would extend only over a limited, conventionally designated, area of
information, "symbolic of the whole institution of privacy".xxxi
According to Fried, this
designated area, whose content may differ considerably from society to society, would
include intimate or sensitive information, and exclude the so-called "public" sphere from its
scope of protection. Fried’s rationale for the "inevitable fact that privacy is gravely
compromised in any concrete social system" is because of "the inevitably and utterly just
exercise of rights by others...".xxxii
For similar reasons, Larry Hunter grants that "although we consider it a violation of
privacy to look in somebody's window and notice what they are doing, we have no problem
with the reverse: someone sitting in his living room looking out his window."xxxiii
Consequently, placing any restraint on such activity would constitute an unacceptable
restraint on liberty—again a manifestation of the "knock down" normative argument.
In the practical arena, as well as in the theoretical realm, public surveillance is
indignantly defended on grounds that it is unreasonable to prevent others from perceiving,
noticing, and talking about the goings-on in public realms. This is form of argument is
favored for protecting the commercial interest in data collection. In the case of Lotus
Marketplace Households, executives defending the proposed product, cited considerations
like these. Denying legal or moral wrongdoing they argued that only information from the
The Problem of Privacy in Public September 28, 2000 13
public domain would be used, no private realms would be breached, and no information
deemed sensitive or intimate would be included.
Versions of the knock-down argument frequently appear in case law. In California
v. Greenwood,xxxiv
for example, which has been cited as a precedent in many subsequent
cases involving (of all things) people’s right to privacy in their garbage, the Supreme Court
ruled that police had not violated the Fourth Amendment when they arranged for
Greenwood’s trash collector to segregate his trash and turn it over to them for inspection.
The court majority offered the following consideration,
Accordingly, having deposited their garbage “in an area particularly suited for
public inspection and, in a manner of speaking, public consumption, for the express
purpose of having strangers take it,” respondents could have no reasonable
expectation of privacy in the inculpatory items that they discarded.xxxv
In another case, United States v. Scott,xxxvi
the court defended the actions of IRS agents,
who had reassembled documents which the defendant had shredded into 5/32-inch strips
before disposing of them in the garbage, arguing,
In our view, shredding garbage and placing it in the public domain subjects it to the
same risks regarding privacy, as engaging in a private conversation in public where
it is subject to the possibility that it may be overheard by other persons. Both are
failed attempts at maintaining privacy whose failure can only be attributed to the
conscious acceptance by the actor of obvious risk factors. In the case of the
conversation, the risk is that conversation in a public area may be overheard by a
third person. In the disposal of trash, the risk is that it may be rummaged through
and deciphered once it leaves the control of the trasher. In both situations the
expectation of privacy has been practically eliminated by the citizen’s own action.
Law enforcement officials are entitled to apply human ingenuity and scientific
advances to collect freely available evidence from the public domain.xxxvii
In Florida v. Riley,xxxviii
this time not involving garbage, the Supreme Court decided
that police had not conducted an illegal search when an officer observed from a helicopter,
at a height of 400 feet, what he thought were marijuana plants. In a separate but concurring
The Problem of Privacy in Public September 28, 2000 14
opinion, Justice O’Connor wrote, “I agree that police observation of the greenhouse in
Riley’s curtilage from a helicopter passing at an altitude of 400 feet did not violate an
expectation of privacy that society is prepared to recognize as ‘reasonable.’xxxix
”xl
She
argued that in the same way it is unreasonable to expect police to shield their eyes so as to
avoid seeing into private property from public thoroughfares, so is it unreasonable for
citizens to expect to be free of aerial observation at altitudes where the “public travel with
sufficient regularity.”xli
In sum, I have tried to show that attempts to define and defend privacy in public,
both in theory and in practice, have been undermined by versions of an argument from
competing interests that I call the normative knock-down argument. It is so named because
it has had a compelling hold over philosophers, policy-makers, and judges, as well as the
commercial interests that benefit from its use.
Empirical
In this section, I outline a third explanation why theorists have seemed to overlook
the problem of privacy in public. I suggest that the divergence of philosophical theory from
popular resentment of surveillance practices is due, in significant measure, to critical
changes which philosophical theory has not yet absorbed because, quite simply, prior to key
developments in information technology, the problem did not exist in a compelling form.
People could count on virtual anonymity even as they traversed the public arena. We see this
assumption at work as the fictional detective, Alexander Gold, interrogate a murder suspect,
"You certainly sounded as though you hated him enough to kill him."
"Not hated, Mr. Gold, despised. If I had killed him, would I have told you how I felt?"
"Maybe. You could be trying reverse psychology."
"Yes, but Professor Moriarty, you know that I know that you really know that I really
know..." Kirsch let his voice fade away.
Alexander had to smile. "All right. Let's talk about something else. Where were you
when Talbott was killed."
"Jogging. In Central Park."
"Witnesses?"
The Problem of Privacy in Public September 28, 2000 15
"Hundreds."…
"So you have an alibi."
"Not exactly. ... "xlii
Seen by hundreds, noticed by none. Most people reasonably make this assumption:
either that they are not noticed, or that any single observer can observe and harbor only
discrete bits of information.xliii
As such, not only would the information be sparse and
disjointed but it would be limited by what any single human brain could reasonably and
efficiently hold. An individual going about his daily activities does not worry about
undue surveillance even if he is observed by one person, on April 4 1997, to be wearing
chinos, a blue polo shirt and loafers and to be tall and blond. By another, he is observed
purchasing three cases of wine from the local liquor store. By a third he is overheard
discussing his son's progress with his school teacher. Later that day, by a fourth, is
observed participating in a march for gay and lesbian rights. All these activities occur in
the public eye; all may be observed, even noted. No single one of these instances of
being observed is necessarily threatening or intrusive.
What has changed? Key advances in computer technology have clearly affected our
facility with information. These advances include an exponential declines in the cost of
computer storage and processing coupled with vast increments in power, the capacity to
create large and complex but decentralized databases on networks of minicomputers and
PCs, the use of expert systems for processing data, and the cooperative handling of data
both within and among institutions.xliv
These developments in information technology and
practices have meant that: a) there is virtually no limit to the amount of information that can
be recorded, b) there is virtually no limit to the scope of analysis that can be done—bounded
only by human ingenuity, and c) the information may be stored virtually forever. These
capabilities combined with alert and intelligent observation have contributed to the practices
and modes of surveillance familiar to us.xlv
The effects of these advances are felt along various dimensions. In the public arena,
not only may the amount of information increase enormously, but information that was once
scattered and transient may now be ordered, systematized, and made permanent. We can do
The Problem of Privacy in Public September 28, 2000 16
things with the information, such as merge and compare real-time observations with past
records, compare those with the records of others, and communicate any of this, at
lightening speed, across networks. Mr. Kirsch would have his alibi, and we would have a
fuller and more systematic picture of the conservatively dressed father protagonist going
about his business on April 4. I discuss the implications of these practices in more detail later
in the paper.
An arena in which these changes have been acutely felt is that of public records.
According to the Freedom of Information Act,xlvi
all governmental records, except those
covered by a specified set of exceptions, including The Privacy Act of 1974, are freely
available for public access. Even though some records of information about people are
covered by The Privacy Act, there are many classes of records with information about
persons, such as birth, death and marriage records, drivers records, real estate ownership
records, court records, and more, that are public. Prior to computerization and advanced
networking capabilities, access to these public records was costly in time and effort. Anyone
seeking information from these records would be required to travel to wherever the records
were housed, such as Courts and Departments of Motor Vehicles, and painstakingly search
for and copy the information they needed. Such effort created de facto protection, serving to
limit access and, therefore, exposure.
The computerization of public records has made them available with far less effort,
either directly from respective government agencies responsible for collecting them, or from
intermediaries who have gathered and organized them. As a consequence, these records are
public in a far more thoroughgoing sense than ever before. In two cases that have come
before the New Jersey Supreme Court, court opinions have acknowledged that the mode by
which information is made public (as in computerized versus paper records) may affect the
actual degree of publicity of these so-called "public" records.xlvii
In a similar vein, those who
have advocated for limiting access to Drivers' Records have argued that when the decision
to allow public access to these records was made, the implications of such records being
public was quite different from what they presently are. In public deliberations, privacy
advocates have suggested that we ought to re-evaluate the meaning of a public record,
including such key issues as the criteria of access to records and the grounds for classifying
The Problem of Privacy in Public September 28, 2000 17
a given database as public. Representatives of other sectors including marketers, information
brokers, and media organizations sharply disagree with such suggestions.xlviii
This important
debate is beyond the scope of this paper.
In review: As a third explanation for neglect of the problem of privacy in public, I
have suggested that until powerful information technologies were applied to the collection
and analysis of information about people, there was no general and systematic threat to
privacy in public. Privacy, as such, was well-enough protected by a combination of
conscious and intentional efforts (including the promulgation of law and moral norms)
abetted by inefficiency. It is not surprising, therefore, that theories were not shaped in
response to the issue of privacy in public; the issue did not yet exist.
III. SHOULD WE PROTECT PRIVACY IN PUBLIC?
To this point, my purpose has been to explain why conceptions of privacy developed by
predominant philosophical and normative theories have not accounted for encroachments on
privacy occurring in so-called "public" realms. For reasons that are conceptual, normative
and empirical in origin, these theories lack mechanisms to deal with conflicts involving
privacy in public and have generally not taken up hard questions about surveillance in nonintimate
realms to determine when such surveillance is morally acceptable and when not.
Implicit in my discussion so far has been an assumption that now bears direct examination,
that normative theories of privacy ought to be concerned with privacy in public, that
contemporary experience with information technology offers compelling reasons to expect
from theory that it provide a means of understanding the problem of privacy in public as
well as a means for adjudicating it.
A prima facie case for caring about public surveillance is that it stirs popular
indignation, worry and resentment. The 30,000 letters of protest against Lotus Marketplace
Households expressed these reactions as do poll results, such as a 1990 poll showing 90%
of respondents agreeing that consumers are being asked to provide excessively personal
information. (57% found it a major problem, 33% a minor problem.)xlix
Individual concerns
The Problem of Privacy in Public September 28, 2000 18
are registered in various ways as shown in the segment below quoted from the RISKS
Forum Digest:
Recently ... several firms have started abusing the power of the Internet to publish
large databases of personal information without permission. This is impolite, and in
many cases it can even be dangerous.
True story: recently, I followed a lead from MacUser magazine to a web
page for dealing with spam e-mailers. That page suggested that one of the first steps
to take was to contact services that track people's e-mail addresses. With growing
horror, I connected to page after page on the list and located myself in their
databases. Some services listed far more than just name and e-mail address. My
home address and phone number were accessible from the same record. Two
services even had a facility to show a map of my neighborhood and the location of
my house in it.
The widespread dispersal of information of this sort, without prior consent,
is a serious invasion of privacy."l
While invectives like this may signal a morally relevant need, they may also be read
as expressions of mere preference, or desire, or even worse, as muddle-headedness. Two
noted contributors to the literature on privacy, William Parent and Tom Gerety, would
explain it as the latter. Both Parent and Gerety assume the burden of sharpening and
clarifying the concept of privacy. Gerety worries that the problem for the concept of
privacy;
comes not from the concept's meagerness but from its amplitude, for it has a protean
capacity to be all things to all lawyers. ... A legal concept will do us little good if it
expands like a gas to fill up the available space.li
While he characterizes privacy as an “island of personal autonomy,”lii
he limits the scope of
this autonomy to the “intimacies of personal identity.”liii
Parent defines a right to privacy
that covers only information that is both personal in nature and not anywhere documented in
The Problem of Privacy in Public September 28, 2000 19
a public place, for example, reported in a newspaper. About all other information, he
concludes that it "cannot without glaring paradox be called private."liv
Thus, for Parent and
Gerety, popular judgment aside, public surveillance would not to be a matter that is covered
by a right to privacy.
I suggest, contrary to approaches like Gerety's and Parent's, that although an
important purpose of philosophical theory is to introduce greater conceptual rigor, a
normative theory that strays too far from ordinary usage and popular sentiment is thereby
rendered unhelpful, or worse, irrelevant. Yet there is still work to be done, for even if we
reject the narrow definitional accounts of theorists like Parent and Gerety, we are not
thereby committed to embracing widespread indignation as, in itself, sufficient reason for
admitting that moral violation has occurred in the activities of public surveillance and data
harvesting. We may regard public expression as a sign, as strongly suggestive, of something
more than preference and mere opinion—more so if it is consistent and fairly widespread—
and we must seek a greater understanding of its source. Only then will we be adequately
guided toward a conclusion about whether privacy in public is a legitimate part of the moral
right to privacy, and if so, under what conditions. To suggest a moral basis for expressions
of popular indignation we must show that popular reaction plumbs human needs that are
deeper and more universal than "mere" preferences and desires.
It is with this purpose that I explore two key aspects public data harvesting. One is
the practice of shifting information from one context to another—usually from the context in
which it was collected, to another context.lv
A second is the set of practices involving
collection, collation, and combination of information drawn from diverse sources in
activities, known variously as "data mining", "profiling", "matching", and the like. Although
the problematic nature of the second set of practices overlaps with first—because it involves
the shifting of information from one context to another—it involves an additional concern,
which I later elaborate. I will argue that these two aspects of public surveillance make
privacy in public an issue which adequate theories of privacy must cover, alongside the
issues that have traditionally been acknowledged as part of their territory.
IV. PRIVACY AND CONTEXTUAL INTEGRITY
The Problem of Privacy in Public September 28, 2000 20
Most people have a robust sense of the information about them that is relevant, appropriate,
or proper to particular circumstances, situations, or relationships. When information is
judged appropriate for a particular situation it usually is readily shared; when appropriate
information is recorded and applied appropriately to a particular circumstance it draws no
objection. People do not object to providing to doctors, for example, the details of their
physical condition, discussing their children's problems with their children's teachers,
divulging financial information to loan officers at banks, sharing with close friends the details
of their romantic relationships. For the myriad transactions, situations and relationships in
which people engage, there are norms—explicit and implicit—governing how much
information and what type of information is fitting for them. Where these norms are
respected I will say that contextual integritylvi
is maintained; where violated, I will say that
contextual integrity has been violated.
Norms governing the appropriateness of information to a context may mark some
information as appropriate for it and some information as inappropriate. It may be
appropriate to expect an employee, for example, to yield a great deal of information to an
employer concerning past employment and education, but inappropriate to have to provide
information about, say, marital status or sexual orientation. Citizens routinely provide a
great deal of information to government agencies and consider it appropriate to do so, but
they are careful about what information they are willing to provide to which agencies. And
there is some information, such as religious affiliation, which they are likely to resist giving
to any government agency at all. Family members know us well, but prying relatives may
rankle us by asking the details of our romantic entanglements. These twinges of indignation
are not necessarily reserved for demands for personal, sensitive, or intimate information.
They occur even when a store clerk requires one's name and address for a cash transaction,
as was standard practice at branches of Radio Shack, or when on-line services ask for
information about one's off-line life, as a subscription to the electronic version of The New
York Times requires of potential subscribers by insisting they complete a questionnaire
asking not only for their names and electronic identification, but also for mailing address,
gender, age, and household income.
The Problem of Privacy in Public September 28, 2000 21
About the norms governing specific relationships and situations, and who determines
these norms—whether by mutual agreement, by authority of one of the participants, through
the shaping influence of culture and society—a great deal could and should be said.
Although I do not here have a ready theory about contexts and the particular norms
associated with them, it is critical to the position on privacy in public that I articulate in this
paper, that such a theory be considered plausible. Furthermore, at least some of the norms
of contextual integrity must be shown to originate from sources other than mere convention,
must be seen as protecting something of independent value to individuals, or to society, or
to both. For if the norms of contextual integrity express only the conventions of the day,
then critics may argue that it is simply a matter of time before people will become
accustomed to the new order brought about by information technology and readily accept
the new privacy conventions of public surveillance. Just as, according to Justice O’Connor,
airplanes have changed the norms of privacy vis-a-vis surveillance from the air, so new
norms will emerge regarding the collection and use of information about persons.
Objections to all the various forms of public surveillance described in the first section of this
paper will cease.lvii
Existing philosophical work on privacy, though it does not address the issue exactly
as defined in the previous paragraph, lends credibility to the idea of independent value
protected by norms of contextual integrity. James Rachels, for example, argues that a right
to privacy ought to include the right not only to control whether information is shared, but
when and with whom it is shared. In having the power to share information discriminately,
people are able to define the nature and degree of intimacy of various relationships:
The same general point can be made about other sorts of human relationships: businessman to
employee, minister to congregant, doctor to patient, husband to wife, parent to child, and so
on. In each case, the sort of relationship that people have to one another involves a conception
of how it is appropriate for them to behave with each other, and what is more, a conception of
the kind and degree of knowledge concerning one another which it is appropriate for them to
have.lviii
The Problem of Privacy in Public September 28, 2000 22
The capacity to define the nature and degree of closeness of relationships is an important
aspect of personal autonomy, Rachels argues, and ought to be protected. Having to enter
relationships or settings with little or no control over what is known about one, may lead to
a sense of having been demeaned, embarrassment, disempowerment, or even fear.
Schoeman sees similar value in respecting norms of contextual integrity. He writes,
People have, and it is important that they maintain, different relationships with
different people. Information appropriate in the context of one relationship may not
be appropriate in another.lix
And elsewhere he illustrates this point,
A person can be active in the gay pride movement in San Francisco, but be private
about her sexual preference vis-a-vis her family and coworkers in Sacramento. A
professor may be highly visible to other gays at the gay bar but discreet about sexual
orientation at the university. Surely the streets and newspapers of San Francisco are
public places as are the gay bars in the quiet university town. Does appearing in
some public settings as a gay activist mean that the person concerned has waived her
rights to civil inattention, to feeling violated if confronted in another setting?lx
People’s judgments that privacy has been violated concur more systematically with
breaches of contextual integrity than with breaches of only intimate or sensitive realms.
Although they may ascribe special status to the latter, they do not thereby accept that
outside of this special realm no norms of privacy apply; they do not accept that outside this
special realm information is detachable from its context and is—we might say—“up for
grabs.” This attitude is reflected in the indignation that may follow as simple a gesture as a
stranger asking a person his or her name in a public square. By contrast, even if information
is quite personal or intimate, people generally do not sense their privacy has been violated
when the information requested is judged relevant to, or appropriate for, a particular setting
or relationship. And this is why traditional theories of privacy, which take as their
guideposts the dichotomy of private versus public, asserting that privacy is morally violated
The Problem of Privacy in Public September 28, 2000 23
only when private information or the private sphere is inappropriately revealed, diverge from
popular judgment which takes contextual integrity as its benchmark. Whereas the former
considers privacy norms as relevant only to private or intimate information, the latter
considers privacy norms as potentially relevant to any information
In the public surveillance currently practiced, information is routinely shifted from
one sphere to another, as when, for example, information about your supermarket purchases
are sold to a list service for magazine subscriptions. At times, the shift may cross not only
contextual lines but temporal lines as information collected in the past—sometimes a very
long time past—is injected into a current setting. (Unlike human memory, which fades,
computer memory lasts indefinitely.)
When the actress Rebecca Shaefer was murdered and police discovered that her
murderer had traced her whereabouts through drivers’ records, people were not only
outraged by the murder but indignant over the means by which her attacker had traced her.
As a result, State Departments of Motor, which have become a fertile source of information
routinely collected from licensed drivers and owners of registered vehicles, have become an
irritant to privacy advocates as well as individuals who are aware of widespread trade in
their computerized records. Public indignation stirred by Shaefer’s murder, and similar
perceived breaches, led to passage of the Drivers' Privacy Protection Act of 1993lxi
which
places some restrictions on the sale of these records. Critics still argue that these restrictions
do not go far enough.lxii
It is commonplace for information deemed not to be “sensitive” to be freely shifted
about, transmitted, exchanged, transferred, and sold. Those who engage in these practices
seem to assume that the information in question has been dislodged from its contextual
attachments and therefore “up for grabs”. Discomfort with the practices involving the
shifting around of information reflects a far different perspective: it suggests that people
judge norms of contextual integrity, and consequently privacy, to have been violated even
when the information in question is not sensitive or intimate. People resent the rampant and
unauthorized distribution of information about themselves not only when they violate the
integrity of an intimate and personal realm, but when they violate contextual integrity. In
The Problem of Privacy in Public September 28, 2000 24
violating contextual integrity they strike at an important aspect of why people care about
privacy.
V. AGGREGATION
At the heart of contemporary data harvesting is the activity known variously as "profiling",
"matching", "data aggregation", and "data mining" in which disparate records, diverse
sources of information about people, are aggregated to produce databases with complex
patterns of information. Smith describes a number of cases. For example, A.T.&T. creating
specialty directories for customerslxiii
based on the aggregated record of their 800 calls;
Citicorp's analyzing the credit card purchases of customers in order to sell profiles to
others;lxiv
banks that Smith studied creating an expert system to categorize individuals into
profile groups by pooling information about them that the banks held; super-bureaus
collecting "information available in many places—from regular credit bureaus (both major
and independent), drivers' license and motor vehicle records, voter registration lists, Social
Security number lists, birth records, court records, etc.,"lxv
in order to devise comprehensive
profiles about individuals that would indicate such things as: purchasing power (credit card
activity index, estimated income, fixed payments, etc.), purchasing activity (active accounts,
bank debits, etc.), shopping data, and demographic data (job, marriage status, dwelling type,
gender, market segment, etc.).lxvi
Data aggregation is by no means limited to the private sector. Used for some time by
law enforcement and the Drug Enforcement Agency, the enterprising San Diego County
government has engaged in the practice for commercial purposes. It created and sold a CDROM
disk containing the aggregated records—including name, address, telephone number,
occupation, birthplace, birthdate, and political affiliation—of 1.25 million of its voters.lxvii
Data subjects and the harvesters of information alike are keenly aware of the
qualitative shift that can occur when individual bits of data are compiled into profiles. From
the perspective of the data gatherers, this capability is one of the most exciting advances that
information technology enables. Institutions in both the public and private sectors, including
law enforcement, financial, and marketing, either take advantage of compiled data directly,
The Problem of Privacy in Public September 28, 2000 25
or buy these products from others—like credit bureaus and list brokers—who specialize in
gathering data from primary sources and organizing it into useful and potentially profitable
forms. Information belies the adage about sewing silk purses out of sows ears, for out of
worthless bits information we may sew assemblages that are rich in value. Assemblages are
valuable for the very reasons that their subjects resent them.
When challenged, supporters and beneficiaries of profiling frequently resort to what I
earlier called the normative "knock down" argument. They argue that there are no good
reasons to prohibit these activities when the information in question is "out there" and
people have made no effort to hide it from view. To prohibit the collection and aggregation
of this information would violate the freedom of those who would observe, record, and
aggregate it. Because the “cat is out of the bag” already, there is no good reason to stifle the
ingenuity of entrepreneurs who would sell and thereby profit from this information. If these
entrepreneurs choose to share what they have learned with others, it would violate their
freedom of expression to stop them.lxviii
Accordingly, any sentiment expressed against
profiling should be treated as such, namely as a sentiment, not as an overriding moral
consideration.
If defenders of aggregation are correct that no private zones are violated, that the
information they use has been provided freely and not under duress, that it is neither stolen
nor leaked, then what could be the privacy interest that is thwarted by the practice of
aggregation?
Even if we grant these defenders of data aggregation their premises, their
conceptions of aggregation—whether sincere or disingenuous—seem to miss something
important about it. It misses whatever element distinguishes the activity of a person casually
looking out his or her window observing the passing scene and the activities described
below in continuation of the paragraph quoted earlier from Hunter's paper:
Consider what happens if I write down everything I see out my window, and all my neighbors
do, too. Suppose we shared notes and compiled the data we got just by looking out our own
windows. When we sorted it all out, we would have detailed personal profiles of everyone we
saw. If every move anyone made in public were recorded, correlated, and analyzed, the veil of
The Problem of Privacy in Public September 28, 2000 26
anonymity protecting us from constant scrutiny would be torn away. Even if that record were
never used, its very existence would certainly change the way we act in public.lxix
The difference between casually observing the passing scene out of one’s window, which
seems perfectly harmless, and the surveillance Hunter has imagined, which seems definitely
sinister, is not merely one of degree. In the passage below, James Boyle in his book,
Shamans, Software, and Spleens, draws attention to a similar concern,
Why do supermarkets offer their preferred customers discounts just for running an
electronic card through a scanner on their way past the checkout? Because
technology now permits the store to keep a precise record of those customers'
purchases and to correlate it with demographic information about them. Advertisers
will soon know everything from our individual brand-name preferences for toilet
paper to the odds that a middle-class family on a particular street will buy Fig
Newtons on a Wednesday. If you are what you eat, then manufacturers will soon
have the information technology to know exactly what you are. This commercially
driven intrusion has not reached Orwellian proportions – at least, not yet.
Nevertheless, information technology has the capacity, if not to end privacy, then to
redefine what we mean by the term.lxx
While the magnitude, detail, thoroughness and scope are important characteristics of
the surveillance described in the two passages, they alone do not account for a sense that a
moral line has been crossed. There are two further considerations that bear mentioning.
First, that the process of compiling and aggregating information almost always involves
shifting information taken from an appropriate context and inserting it into one perceived
not to be so. That is, the violation of contextual integrity is part of the reason critics find
data aggregation to be morally offensive. A second consideration, striking closer to the core
of the practice of profiling, is that while isolated bits of information (as generated, for
example, by merely walking around in public spaces and not taking active steps to avoid
notice) are not especially revealing, assemblages are capable of exposing people quite
profoundly.
The Problem of Privacy in Public September 28, 2000 27
The value of aggregates is that they are multidimensional and as such provide more
information than pictures that are less filled out. Beyond this, however, an aggregate can
incorporate a richer portrait of the individual than even the bits taken together (i.e. the
whole being more than the sum of parts) as it may include not only information explicitly
given but information inferred from that which has been given. As Jeffrey Reiman observes,
...by accumulating a lot of disparate pieces of public information, you can construct a
fairly detailed picture of a person's private life. You can find out who her friends are,
what she does for fun or profit, and from such facts others can be inferred, whether she is
punctual, whether she is faithful, and so on.lxxi
If we know, for example, that someone has purchased a home pregnancy test, we can infer
with some degree of certainty the nature of activities in which she has recently engaged; if a
person has registered as a Republican we can infer with some degree of certainty how he or
she would react to a range of social and political issues; if someone owns a house in affluent
Palo Alto, we can infer his or her minimum financial holdings. In other words, portraits may
provide descriptive access to an individual, multiple forms of identification, and a sense of
what they care about.
The picture of a person that a profile provides can, for the reasons given, be broad,
deep and traverse time. These pictures may be rich enough to reveal aspects of an
individual's character, to ground predictions about their propensities, and even suggest ways
of manipulating them. One provider of such a service boasts as follows:
With a 98% compliance rate, our registered users provide us with specific
information about themselves, such as their age, income, gender and zip code. And
because each and every one of our users have verifiable e-mail addresses, we know
their data is accurate -- far more accurate than any cookie-based counting.
Plus, all of our user information is warehoused in a sophisticated database,
so the information is stable, accessible and flexible.
The Problem of Privacy in Public September 28, 2000 28
Depending on your needs, we can customize user groups and adjust
messages to specific segments, using third-party data or additional user-supplied
information. So you can expand your targeting possibilities.
What's more, because they're New York Times on the Web subscribers, our
users are affluent, influential and highly engaged in our site.lxxii
Demographic profiles, financial profiles, and consumer profiles identify people as suitable
targets for proposed "treatments." Used in this way, a profile may be seen as a device that
offers a wayof targeting people as the likely means to fulfilling someone else's end.
In sum, the two preceding sections argue that the negative reactions to public
surveillance are due at least in part to characteristics of public surveillance that are genuinely
morally objectionable. One is that public surveillance practices regularly violate norms of
contextual integrity when information readily revealed in one context, and public with
respect to it, is transmitted to, and revealed in, another. The importance of integrity of
contexts, which has been recognized in relation to intimate and sensitive realms, has not
been sufficiently acknowledged in other realms. Also morally objectionable are the activities
integral to public surveillance practices known as profiling, data aggregation, and data
mining, which provide the means to reach, target, and possibly manipulate their subjects.
VI. PRIVACY IN PUBLIC: A GENUINE PRIVACY INTEREST
I began this paper by suggesting that philosophical theories of privacy, in responding
primarily to the threat of governmental intrusion into privacy and to the threat of any
intrusions into the personal, intimate realms, fail to respond to an important and growing
challenge to privacy. My purpose has been to argue that public surveillance, which many
theorists have denied a central place, ought often to be construed as a violation of genuine
privacy interests. Although I have criticized predominant theories of privacy for neglecting
this important privacy interest, I rely on the considerable insights developed in these theories
to show why even in the public sphere individuals have a legitimate privacy interests. It also
remains for the courts as well as further theoretical work to develop criteria for
The Problem of Privacy in Public September 28, 2000 29
distinguishing between those acts of public surveillance that seem not to violate privacy and
those that do.
Among the essential contributions that these theories make is drawing the
connection between privacy and other values. For many, privacy is valuable, is worth
protecting as either a moral or legal right, or both, because it functions to protect and
promote other important ends.lxxiii
Alan Westin, for example, in his influential book Privacy
and Freedom, asserts that privacy promotes important human ends in a democratic, free
society: it enhances personal autonomy (which he understands as "the desire to avoid being
manipulated or dominated wholly by others"lxxiv
), it creates a protected realm for emotional
release, provides a context in which an individual can "exert his individuality on events",lxxv
and the creates the possibility of limited and protected communication. Ruth Gavison offers
another persuasive account of the essential role privacy plays in safeguarding or promoting
other deeply held values including liberty of action, "mental health, autonomy, growth,
creativity, and the capacity to form and create meaningful human relations".lxxvi
Several
other exemplary works on privacy offer analogous insights, demonstrating the value of
privacy both for individuals and society. Although I articulate my analysis in terminology
drawn primarily from Westin and Gavison, it is not necessarily tied to the details of their
theories.
These approaches have in common a version of the idea that privacy protects a "safe
haven", or sanctuary, where people may be free from the scrutiny and possibly the
disapprobation of others. Within these private spheres people are able to control the terms
under which they live their lives.lxxvii
By exercising control over intimate and sensitive
information about themselves, people may exercise control over the way they portray
themselves to others, especially those others with whom they engage in lasting relationships.
These two forms of privacy, namely, control over information and control over access,
establish the conditions for a free society and, among other things, enhance people’s
capacity to function as autonomous, creative, free agents.
In the world before powerful computers, virtually limitless storage capacity,
software for information management, and network capabilities, privacy was well enough
protected by safeguarding sensitive information and intimate spheres against unwanted
The Problem of Privacy in Public September 28, 2000 30
intrusion. Through a relatively narrow range of prohibitions, privacy was afforded a decent
level of protection because, as discussed in the section on empirical reasons for the neglect
of privacy in public, the prohibitions themselves were abetted by conditions such as the
limits of human memory, polite indifference, and inconvenience.
But these conditions no longer hold. In their place we have powerful information
technology coupled with an insatiable desire to know—whatever now may be useful to
someone, somewhere, or what may become so in the future. Information is fluid and
comprehensive; cleverly devised profiles constitute a powerful tool for understanding
people, influencing their behavior, and even manipulating them. Those who are not fully
aware what or how much others know about them are more easily targeted or manipulated.
Those with greater awareness and understanding may be able to protect their privacy more
effectively, but at the expense of developing a wariness, self-consciousness, suspicion, and
even tentativeness in their relations with others. DeCew describes this as “a chilling effect”
on behavior.lxxviii
The values that were once relatively well shielded through the fortification
of the intimate realm are now vulnerable via other, supposedly public, approaches. Because
there is more at stake in an individual's controlling even non-intimate information, it no
longer self-evident that the balance must favor of the freedom of those who seek to observe
and record when weighed against the privacy interests of those who are observed.
These considerations support the view that popular reaction to public surveillance is
not merely a reflection of popular—possibly irrational—sentiment but a recognition that
prominent elements of public surveillance constitute a genuine moral violation of privacy.
Reasons for protecting privacy in public are quite similar to reasons for protecting privacy of
the more traditional kind because values placed in jeopardy from invasions of the intimate
realm are also jeopardized by various forms of public surveillance practiced today. As noted
earlier, these values are wide-ranging, including individual values such as autonomy, liberty,
individuality, capacity to form and maintain intimate relations, mental health, creativity,
personal growth; as well as social values such as a free and democratic society. Those who
engage in contemporary practices of public surveillance have discovered a novel way to
eavesdrop, to spy on, to learn more about people than they have a legitimate right to know.
The Problem of Privacy in Public September 28, 2000 31
And preventing this constellation of intrusions is one of the fundamental protections that
privacy offers.
VII. IMPLICATIONS FOR POLICY
The purpose of this paper has been to present a case for extending or revising existing
philosophical theory, or developing new theories, that would accommodate privacy in
public. I hope to have succeeded in this. Although the purpose has not been to recommend
or craft specific privacy policies, I would like, in these concluding paragraphs, to consider
whether a recognized interest in privacy in public could have any power to shift the course
of privacy policy in the United States, which at present, gives no systematic consideration to
it.
I see two means. One would be to emphasize the principle of contextual integrity in
order to weaken the influence of the private-public dichotomy in setting the agenda for
privacy policy, as well as theory. The idea of contextual integrity and the norms emerging
from it ought not be utterly foreign. There is, after all ample precedent in relationships that
explicitly call for confidentiality such as, physician to patient, clergyman to congregant, and
so on. We can view these relationships and contexts that call for confidentiality as instances
of a more general requirements of contextual integrity. We may likewise view the Video
Privacy Protection Act as giving legal protection to the video rental context, also an
extension of the familiar professional settings. Building upon such cases, we might extend
application of a principle of contextual integrity further to cover various settings such as
medical insurance bureaus, charitable organizations to which one has donated, some as
mundane as supermarkets, and more.lxxix
Some privacy advocates object to the approach just described—a “sectoral”
approach—favoring a second, “omnibus” approach. This second approach accords a strong,
comprehensive right to privacy which grants control to individuals over all information
about themselves irrespective of context. The European Union’s privacy initiative,
scheduled to take effect in 1998, is considered an example of this approach.lxxx
Recognizing
a fundamental right to privacy shifts the burden away from individuals having to
The Problem of Privacy in Public September 28, 2000 32
demonstrate the importance of maintaining control over various especially sensitive
categories of information onto potential gatherers and users of information, who would need
to demonstrate a critical need for the information. Although it is important to show that
there are practical feasible policy mechanisms for protecting privacy in public, I will not
pursue the details of these options here.
Before concluding, I will consider a possible objection to the protection of privacy in
public, namely the objection I earlier called, the normative "knock-down" argument. Are we
in a position to better understand this argument, and more important, will we be able to
defend privacy in public against it?
As we have seen, those who invoke a normative knock-down argument against
protecting privacy in public usually point out that the information in question is neither
intimate nor sensitive. They also say that because the information in question has been freely
exposed in public by its subjects, it is unreasonable and wrong for their subjects to claim a
right to prevent access to it or use of it.
In responding to such an argument I would suggest, first, that some of its power is
based on an equivocation on the “it” to which subjects have supposedly given implicit
consent. While shoppers in a supermarket have implicitly consented to fellow shoppers
seeing the contents of their shopping carts—they do not expect fellow shoppers not to
look—they have neither implicitly nor explicitly agreed to others collecting the information
and selling it to third, fourth, etc. parties so that the data may be warehoused, mined, and
assembled, so that their behavior may be modeled and manipulated. Just as someone buying
a pregnancy test in a drugstore may have not choice but to expose this bit of information to
fellow shoppers, they have not thereby acceded to unrelenting publication of their sexual
behavior.
A detractor may still balk. To incorporate protection for privacy in public into law
and public policy is, nevertheless both unrealistic and unreasonable. Even if the moral
authority of the normative "knock-down" argument has been undermined, its practical force
remains evident in Reiman's warning. The challenge remains that if one is willing to be open,
and behave openly, it would be an oppressive society that enforced norms of privacy that
entailed a right never to be seen on a crowded street. The burden placed on others cannot
The Problem of Privacy in Public September 28, 2000 33
interfere with the normal activities of their daily lives, we cannot expect in general, as Justice
O’Connor wrote about police officers in particular, that people “shield their eyes when
passing by.”lxxxi
Although it seems both impossible and wrong to impose so great a burden on people
in order to protect privacy in public, it is not impossible to articulate other measures of
protection that are not overly burdensome and at the same time do not unduly compromise
what is valuable in privacy in public. This would involve recognizing the distinction between
exposing something for observation, on the one hand, and yielding control over it, on the
other. Although at first this may seem practically difficult or even impossible, a model for
policy based in recognizing such a distinction may be found in another area of discourse --
intellectual property. Two central mechanisms for protecting intellectual property, patent
and copyright, are devised expressly for the purpose of allowing something to be exposed
(in this case, the works of intellectual labor) without yielding control over it. While I do not
support the position, sometimes put forward, of privacy as a form of self-ownershiplxxxii
(a
debate for another occasion), I suggest that for purposes of crafting reasonable policy, the
practical mechanisms developed in the service of intellectual ownership, which are socially
entrenched and for the most part successful, may serve well for the purpose of protecting
privacy.
This paper has argued for a right to privacy that would encompass privacy in public.
Although it does not articulate a theory from which this extended right can be derived, it has
advanced principles to guide the development of such a theory, principles according to
which activities that, in the past, have fallen outside the scope of many influential legal and
philosophical theories, may be judged relevant to a moral right to privacy. I have in mind,
here, the principle of contextual integrity and the principle that no information is genuinely
“up for grabs”, available for purposes such as aggregation, profiling, and data mining.
These principles offer criteria for discriminating from among the various forms of public
surveillance and record-keeping those that constitute moral violations of privacy and those
that do not.
The Problem of Privacy in Public September 28, 2000 34
i
I am grateful to many colleagues who generously contributed to this paper with excellent comments
and suggestions: Phil Agre, Judith Wagner DeCew, Jodi Halpern, David Heyd, Jerry Kang, John
Kleinig, Gary Marx, David Orentlicher, Kristen Shrader-Frechette, Jeroen van den Hoven, and Tom
Vogt. I am also indebted to anonymous reviewers for Law and Philosophy for careful reading and
several wise suggestions.
.
ii
Anita Allen recently drew my attention to a discussion in Allen, A., Uneasy Access (Totowa, New
Jersey: Rowman & Littlefield, 1988), Chapter 5, in which she discusses whether, and when, it is
reasonable to expect that privacy will be respected in public spaces. She argues that even in public
places like hiking trails, subway cars, or bars, people ought to be free of invasive surveillance. She
also considers sexual harassment in public spaces and the public display of pornography to be
activities that violate privacy in public. Also, see Helen Nissenbaum, “Toward an approach to
privacy in public: the challenges of information technology,” Ethics and Behavior, 7, no. 3 (1997):
207-219, where I introduce the concept of privacy in public.
iii
Larry Hunter, “Public Image,” Whole Earth Review (January, 1985). Reprinted in Deborah
Johnson and Helen Nissenbaum Computers, Ethics, and Social Values (Englewood Cliffs: Prentice
Hall, 1995), p. 294.
iv
In their various writings but see, especially, Judith Wagner DeCew, In Pursuit of Privacy: Law,
Ethics, and the Rise of Technology (Ithaca: Cornell University Press, 1997) and Ferdinand
Schoeman, Privacy and Social Freedom (Cambridge: Cambridge University Press, 1992).
vv
For example, see “Goals Clash in Shielding Privacy,” The New York Times, October 28, 1997, “In
Prison, Free to Get Information,” The New York Times, October 20, 1997, “On Line, High-Tech
Sleuths Find Private Facts,” The New York Times, September 15, 1997, No More Privacy: All About
You, Films for the Humanities and Sciences, Inc., Princeton: 1993. “The Death of Privacy,” Time,
August 25, 1997, Col. 150 No. 8.
vi
H. Jeff Smith, in Managing Privacy: Information Technology and Corporate America (Chapel
Hill: The University of North Carolina Press, 1994), reports that post offices release lists to owners
of target marketing consisting of the names and addresses of individuals who complete National
Change of Address cards.
vii
Molecular biologists predict that one day, in the not too distant future, a computer chip will be
capable of recording each individual’s complete DNA sequence in something analogous to a bar-
code.
viii
One of the devices for doing so, affectionately called "cookies," is coming under fire from
perspectives both of security and privacy. Cookies are small programs that are transmitted from one
site to another (usually from a web page to a web browser) for the purpose of conveying information
about a user's system configuration, usage information, as well as other information that a user
voluntarily provides to the cookie.
ix
Partly because of this, the battle over encryption is so hard fought, with privacy advocates arguing
that access to the full capabilities of encryption should be available to individuals.
x
For example, TRW Credit Data, Equifax and Trans Union, the three major (super) credit bureaus.
The Problem of Privacy in Public September 28, 2000 35
xi
See Iver Peterson, "Public Information, Business Rates: State Agencies Turn Data Base Record
Into Cash Cows," The New York Times, July 14, 1997.
xii
Higg-A-Rella, Inc. v. County of Essex; 141 N.J. 35 (1985).
xiii
In, Managing Privacy, H. Jeff Smith describes a parallel situation in the business world where
corporate policy on privacy is fragmented and not always internally consistent. One company’s
privacy policies, usually devised in isolation from those of other companies, may differ enormously
in what they allow and disallow with the information they gather. They frequently do not admit to
being driven by any underlying “right to privacy,” but prefer to portray their policies as being driven
by prudence and public perception. Corporations continue to resist public policy that would impose
governmental regulation on their use of information about persons.
xv
In legal terms, what may be referred to as a person’s “curtilage”.
xvi
The other two are TRW and Trans Union Credit Information.
xvii
Other industry analysts were also very encouraging. An interesting example is Esther Dyson, now
head of the Electronic Frontier Foundation, in “Data is Dandy,” Forbes (April, 1990), p. 180.
xviii
Helen Nissenbaum, "Toward an Approach to Privacy in Public: Challenges of Information
Technology".
xix
I should qualify. First, there are elements in existing theories, even those that do not directly
address the problem of privacy in public, that I will show are relevant to it. Second several writers
have written about privacy in ways that overlap with my concern with “privacy in public.” As
mentioned earlier, these include Ferdinand Schoeman and Judith DeCew. Specific references to their
works are given in subsequent footnotes.
xx
I do not mean to suggest that there is universal agreement among scholars either about the
strictness of the dichotomy or the meaning of the respective concepts. Stanley J. Benn and Gerald F.
Gauss (Eds.), Public and Private in Social Life, (London and Canberra: St. Martin’s Press, 1983),
suggest that although the concepts of private and public serve to organize norms of access, agency
and interest, the dichotomy is not as clear and consistent as some would have us believe. For
example, a context can be conceived as both public and private: a living room in a house is
considered private in relation to the outside, but public in relation to bedrooms in the house.
xxi
Judith DeCew, In Pursuit of Privacy: Law, Ethics, and the Rise of Technology, p. 10.
xxii
As David Heyd pointed out to me, the word “statistics” is derived from the word “state”.
Government involvement in the practice of collecting information about populations, such as in
census-taking, goes back many centuries, and even discussed in the Bible. Alan Westin, Privacy and
Freedom (New York: Atheneum, 1965), Priscilla Regan, Legislating Privacy, and Kenneth Laudon,
Dossier Society: Value Choices in the Design of National Information Systems (New York:
Columbia University Press, 1986) all discuss aspects of privacy protection against government
intrusion.
The Problem of Privacy in Public September 28, 2000 36
xxiii
See Priscilla Regan, Legislating Privacy: Technology, Social Values, and Public Policy, for an
excellent discussion of privacy policy. Regan pinpoints the SSRC’s 1965 proposal as a key point in
the history of privacy policy with respect to records of information about people.
xxiv
Julie Innes in her book, Privacy, Intimacy and Isolation (New York and Oxford: Oxford
University Press, 1992) articulates one such view of privacy in which intimacy is a defining
characteristic. Also, see Nissenbaum, “An Approach to Privacy in Public,” for a fuller discussion of
approaches to privacy that have focused on privacy as a protection for the intimate realm.
xxv
Jeffrey Reiman, "Privacy, Intimacy and Personhood,” Philosophy & Public Affairs 6, no. 1 (Fall
1976): 26-44.
xxvi
Reiman, “Privacy, Intimacy and Personhood,” pp. 43-44.
xxvii
Reiman, “Privacy, Intimacy and Personhood,” p. 44.
xxviii
The idea behind trade secrets is similar. A secret earns legal protection only if owners take
adequate measures to keep it out of the public eye.
xxix
Again, Schoeman, discussed later, is a notable exception. Also see Jeffrey Reiman, “Driving to the
Panopticon: A Philosophical Exploration of the Risks to Privacy Posed by the Highway Technology
of the Future," Santa Clara Computer and High Technology Law Journal, (Volume 11, Number 1,
March 1995).
xxx
Charles Fried "Privacy,” The Yale Law Journal (Volume 77):475-93. p. 493/
xxxi
Fried, “Privacy,” pp. 488-489.
xxxii
Fried, “Privacy,” p. 487.
xxxiii
Larry Hunter, “Public Image” p. 295.
xxxiv
486 U.S. 35, 108 S.Ct. 1625, 100 L.Ed.2d 30 (1988).
xxxv
This case and a series of related cases are discussed in LaFave, W.R. Search and Seizure: A
Treatise on the Fourth Amendment, Third Edition, Volume 1, (St. Paul, Minn.: West Publishing
Co., 1996).
xxxvi
975 F.2d 927 (1st Circ.1992).
xxxvii
Quoted from LaFave, Search and Seizure, p. 603.
xxxviii
488 U.S. 445; 109 S. Ct. 693; 1989 U.S. LEXIS 580; 102 L. Ed. 2d 835; 57 U.S.L.W. 4126.
xxxix
Katz v. United States, 389 U.S. 347, 361 (1967).
xl
Florida v. Reilly (O’Connor, J. concurring).
The Problem of Privacy in Public September 28, 2000 37
xli
One of the anonymous reviewers for Law and Philosophy points out that newly developed method
of government surveillance, for example, through walls, makes this issue even more pressing and
problematic.
xlii
Herbert Resnicow, The Gold Solution (New York: St. Martin's Press, 1983), pp. 116-117.
xliii
I exclude here special cases such as when suspects are surveilled by law enforcement officers with
a special purpose, such as, hoping to catch them in the act of purchasing a shipment of heroin.
xliv
I draw on H.M. Deitels' characterization of the period of 1970s to the present, which he describes
as the "Fourth Generation of information technology." This is discussed in Smith, Managing
Privacy, pp. 180-181.
xlv
James Rule also credits changes in social organization, now driven by large anonymous
institutions, along with peoples’ desire to be treated as individuals. See James Rule, et. al.,
“Preserving Individual Autonomy in an Information-Oriented Society,” in Computer Privacy in the
Next Decade, Lance Hoffman (ed.), (New York: Atheneum: 1980), 65-87.
xlvi
F.O.I.A. 5 U.S. Code, sec 552, 1966, strengthened in 1974 and 1976.
xlvii
See Higg-A-Rella, Inc. v. County of Essex. 141 N.J. 35 (1985) and Doe V. Poritz, 142 N.J. 1
(1995), discussed in greater detail in Nissenbaum, "Toward and Approach to Privacy in Public:
Challenges of Information Technology," op. cit.
xlviii
For example, as debated at the Public Hearings of the Information Task Force Information Policy
Committee Working Group on Privacy, held on January 26-27, 1994, Washington DC.
xlix
Smith, Managing Privacy, p. 125.
l
Jon Handler, submitted to RISKS Forum Digest, DEC 23, 1996. RISKS is a moderated bulletin
board whose purpose is to publicize and resolve computer-related risks. It is held in high regard
within the community of security experts and software engineers.
li
Tom Gerety, "Redefining Privacy," Harvard Civil Rights-Civil Liberties Law Review, vol. 12 No.
2 (1977): 233-293, p. 234.
lii
Gerety, “Redefining Privacy,” p. 271.
liii
Gerety, “Redefining Privacy,” p. 281.
liv
William Parent, "Privacy, Morality, and the Law," Philosophy & Public Affairs Vol. 12 no 5
(1983): 269-288, p. 271. See also DeCew, In Pursuit of Privacy: Law, Ethics, and the Rise of
Technology, especially Chapter 2, for a careful critique of Parent’s position.
lv
A term for this that has entered the vocabulary of on-line discussions is “data creeping.”
lvi
A similar idea has been proposed by the philosopher, Jeroen van den Hoven. He uses the term,
“spheres of access,” to cover essentially the same idea as “contextual integrity.”
The Problem of Privacy in Public September 28, 2000 38
lvii
I am grateful to Philip Agre for prodding me into seeing that simply asserting the presence of
norms is not grounds enough for rejecting a new practice that violates the norms. We need further to
show that the norms are more than “mere” convention and that they protect something of genuine
value to individuals or society or both. Ferdinand Schoeman, in Privacy and Social Freedom
(Cambridge: Cambridge University Press, 1992), Chapter 10, introduces a similar concept in his
discussion and literary exploration of “spheres of life.”
lviii
James Rachels, "Why Privacy is Important,” Philosophy & Public Affairs 4, no. 4 (Summer
1975): 323-333, p. 328.
lix
Ferdinand Schoeman, "Privacy and Intimate Information," in F. Schoeman (ed.) Philosophical
Dimensions of Privacy: An Anthology, (Cambridge: Cambridge University Press, 1984), p.408.
lx
p.73 Schoeman, F. "Gossip and Privacy" in R.F. Goodman and A.B. Ze’ev (eds.) Good Gossip,
(University Press of Kansas, 1994) 72-84.
lxi
103rd Congress, H.R. 3365.
lxii
In another driver-related case, privacy advocates worry about E-ZPass, the electronic toll system
operating on toll roads and bridges in the East Coast of the United States, operated by the
Triborough Bridge and Tunnel Authority. Electronic devices installed in a motor vehicle transmit
information about identity for billing purposes. Critics worry that information about drivers’
whereabouts may be used in unrelated contexts. Apparently, the New York Police Department
successfully fought against a requirement that records be closed to access except via subpoenas.
lxiii
Smith, Managing Privacy, p.185.
lxiv
Smith, Managing Privacy, p. 186.
lxv
Smith, Managing Privacy, p. 124.
lxvi
Smith, Managing Privacy, pp. 114-115.
lxvii
Smith, Managing Privacy, p. 190.
lxviii
This sort of rhetoric was present during the Lotus Marketplace incident.
lxix
Hunter, L. "Public Image," p. 295.
lxx
James Boyle, Shamans, Software, and Spleens, (Cambridge: Harvard University Press, 1996) p. 3-
4.
lxxi
Jeffrey H. Reiman, "Driving to the Panopticon: A Philosophical Exploration of the Risks to
Privacy Posed by the Highway Technology of the Future," Santa Clara Computer and High
Technology Law Journal, Volume 11, Number 1 (March 1995).
lxxii
Advertisement, The New York Times, Monday July 14, 1997.
The Problem of Privacy in Public September 28, 2000 39
lxxiii
This claim is not incompatible with the stronger claims that some make about privacy, that it is
valuable not only because it is instrumental in achieving other ends but as an end in itself.
lxxiv
p.33, Alan F. Westin, Privacy and Freedom, New York: Atheneum, 1967.
lxxv
p.36, ibid.
lxxvi
Ruth Gavison, “Privacy and the Limits of the Law,” p. 442.
lxxvii
John Kleinig reminds me that even these freedoms are limited. One cannot, for example, claim
protection for spousal abuse on grounds that it occurs in private.
lxxviii
Judith Wagner DeCew, In Pursuit of Privacy: Law, Ethics, and the Rise of Technology,
(Ithaca: Cornell University Press, 1997) p. 64.
lxxix
This takes us into the territory of hard-fought battles over whether integrity would be more aptly
protected through opt-in versus opt-out. Opt-in, in my view, is far truer to privacy requirements but I
will not take up the matter here.
lxxx
This is the approach that is incorporated in the European Union Privacy Directive which is
scheduled to take effect in 1998.
lxxxi
488 U.S. 445; 109 S. Ct. 693; 1989 U.S. LEXIS 580; 102 L. Ed. 2d 835; 57 U.S.L.W. 4126,
(O’Connor, J., concurring).
lxxxii
See for example Laudon, K. "Markets and Privacy," Communications of the ACM, September,
Vol. 39, No 9. (1996).