Perhaps it is fitting that the topic of surveillance, so shrouded in intrigue, is often ambiguous and misunderstood. One error involves the common tendency to see surveillance as the opposite of privacy. Another is to associate it only with government and with law and order activities in particular. Privacy and surveillance can be interwoven. Viewed in social process terms, they are different sides of the same coin. Surveillance may be the means of crossing borders that protect privacy. This is illustrated by strong public concern with the nefarious and multifarious electronic, chemical, and database means of collecting personally identifiable information without the individual’s knowledge and consent. Yet surveillance can also be the means of protecting privacy. Consider the passwords and audit trails required to use some databases, or the various defensive measures such as a perimeter video camera used to protect the home. Whether surveillance or privacy invasion is seen depends partly on the point of view taken.
When this duality is considered in the abstract, it is difficult to reach broad conclusions about the appropriateness of either surveillance or privacy. They can be socially desirable, as well as destructive. Surveillance can serve goals of protection, administration, rule compliance, documentation, and strategy, as well as goals involving inappropriate manipulation, restricted life opportunities, social control, and spying. Privacy can be central to an individual’s dignity and liberty as well as to intimacy, honest communication, group borders, and democracy, but it can also hide socially destructive behaviors such as abuse within families and white collar crime.
With the development of new communication and computer technologies and new ways of living that increasingly rely on non-face-to-face forms of interaction, questions of personal information have taken on new social significance. The microchip and computer are central to surveillance developments and, in turn, reflect broader social forces set in motion with industrialization that involve empirical documentation, rationality, bureaucracy, and capitalism. The increased availability of personal information is one strand in the constant expansion of knowledge witnessed in recent centuries, and of the centrality of information to the workings of contemporary society.
The dictionary defines surveillance as “close observation, especially of a suspected person” or persons. Examples include an individual suspected of bank robbery who is discretely followed by police and is apprehended after robbing another bank, or the discovery that a leader of an anti-globalization protest movement is a police informer.
The above examples are instances of “traditional surveillance,” and the dictionary definition fits. Yet it is too narrow. The focus of surveillance goes beyond suspects, crime, and national security. To varying degrees, surveillance is a property of any social system—from two friends to a workplace to a government. Examples of this understanding of surveillance include a supervisor monitoring an employee’s productivity, a doctor assessing the health of a patient, a parent observing a child at play in the park, the driver of a speeding car asked to show a driver’s license, or a voyeur. Each of these also involves surveillance.
Information boundaries and contests are found in all societies and in all living systems. Humans are curious and yet also seek to protect their informational borders, even as they must also reveal their information. To survive, individuals and groups engage in, and guard against, surveillance. Seeking information about others, whether within or beyond one’s group, is characteristic of all societies, as are efforts to protect information. However, the form, content, and rules about information vary considerably. In the case of surveillance, for example, there are variations between relying on informers, intercepting smoke signals, taking satellite photographs, gathering information from “cookies” placed on the computers of Internet users, or mapping the spread of a contagious disease. With respect to communicating information, there are differences between the expectation that close friends will share secrets, the varied disclosure notice and Freedom of Information Act requirements, and the expectations surrounding the confidential-ity of sealed or classified records.
The traditional forms of surveillance contrast in important ways with what can be called the “new surveillance,” a form that became increasingly prominent toward the end of the twentieth century. The new social surveillance can be defined as “scrutiny through the use of technical means to extract or create personal or group data, whether from individuals or contexts.” Examples include the use of video cameras; computer matching, profiling and data mining; work, computer, and electronic location monitoring; DNA analysis; drug testing; brain scans for lie detection; various self-administered tests; and thermal and other forms of imaging to reveal what is behind walls and enclosures. The use of “technical means” to extract and create the information implies the ability to go beyond what is offered to the unaided senses or voluntarily reported. Much new surveillance involves an automated process that extends the senses and cognitive abilities through using material artifacts or software.
Using the broader verb “scrutinize” rather than “observe” in the definition calls attention to the fact that contemporary forms of surveillance often go beyond the visual image to involve sound, smell, motion, numbers, and words. The eyes do contain the vast majority of the body’s sense receptors, and the visual is a master metaphor for the other senses (e.g., saying “I see” for “I understand”). Yet the eye as the major means of direct surveillance is increasingly joined or replaced by other means. The use of multiple senses and sources of data is an important characteristic of much of the new surveillance.
Traditionally, surveillance involved close observation by a person, not a machine. But with contemporary practices, surveillance may be carried out from afar, as with satellite imaging or the remote monitoring of communications and work. Nor need it be done through close, detailed monitoring—much initial surveillance involves superficial scans looking for patterns of interest to be pursued later in greater detail. Surveillance has become both farther away and closer than previously. It occurs with sponge-like absorbency and laser-like specificity.
In a striking innovation, surveillance has also been applied to contexts (geo-graphical places and spaces, particular time periods, networks, systems, and categories of person), not just to a particular person whose identity is known beforehand. Moreover, the new surveillance technologies are often applied categorically (e.g., all employees are drug tested or all travelers are searched, rather than those whom there is some reason to suspect).
Traditional surveillance often implied a noncooperative relationship and a clear distinction between the object of surveillance and the person carrying it out. In an age of servants listening behind closed doors, binoculars, and telegraph interceptions that separation made sense. It was easy to distinguish the watcher from the person being watched. In the new surveillance, with its expanded forms of self-surveillance and cooperative surveillance, the easy distinction between the agent and the subject of surveillance can be blurred.
In related forms, subjects may willingly cooperate by submitting to personal surveillance in order to have consumer benefits (e.g., frequent flyer and shopper discounts) or for convenience (e.g., access to fast-track lanes on toll roads for which fees are paid in advance). Implanted chips transmitting identity and location that were initially offered for pets are now available for their owners (and others) as well. In some work settings, smart badges worn by individuals do the same thing, although not with the same degree of voluntarism.
Beyond the individual forms, surveillance at an aggregate level is central to social management. The careful tracking of behavior through computer records (referred to by Roger Clarke as “dataveillance”) is believed to offer a more rational and efficient approach to social organization and control. The new surveillance relative to traditional surveillance has low visibility, or is often invisible. Manipulation as against direct coercion has become more prominent. Monitoring may be purposefully disguised as with a video camera hidden in a teddy bear or a clock. Or it may simply come to be routinized and taken for granted as data collection is inte-grated into everyday activities. For example, use of a credit card for purchases automatically conveys information about consumption, time, and location of the purchase.
With the trend toward ubiquitous computing, surveillance and sensors in one sense disappear into ordinary activities and objects, automobiles, cellular téléphones, toilets, buildings, clothes, and even bodies. The relatively labor-intensive bar code on consumer goods that requires manual scanning may soon be replaced with inexpensive embedded radio frequency identification (RFID) computer chips that can be read automatically from short distances. The remote sensing of preferences and behavior offers many advantages, such as controlling temperature and lighting in a room or reducing shipping and merchandising costs, while also generating records that can be used for surveillance.
There may be only a short interval between the discovery of the information and the automatic taking of action. The individual as a subject of data collection and analysis may also become almost simultaneously the object of an intervention, whether this involves the triggering of an alarm or the granting (or denial) of some form of access, for example, the ability to enter through a locked door, use a computer, or make a purchase.
The new forms are relatively inexpensive per unit of data collected. Relative to traditional forms, it is easy to combine visual, auditory, text, and numerical data. It is relatively easier to organize, store, retrieve, analyze, send, and receive data. Data are available in real time, and data collection can be continuous and offer information on the past, present, and future, as through statistical predictions. Simulated models of behavior can then be created. The new surveillance is more comprehensive, intensive, and extensive. The ratio of what the individual knows about himself or herself relative to what the surveilling organization knows is changing dramatically.
I. Surveillance Structures and Processes
Surveillance can be analyzed by breaking it into components. These indicate where to look and what to measure. This can help in identifying the variation that is central to explanation, understanding, and evaluation. Structures that are fixed at one point in time (like a photograph) can be differentiated from processes that involve interaction and developments over time (like a video). The next section considers some surveillance structures.
Organizational surveillance is distinct from the non-organizational surveillance carried on by individuals. As James Rules has noted, modern organizations are the driving forces in the instrumental collection of personal data. As organizations increasingly use personal data for what David Lyon calls social sorting, the implications for many aspects of life are profound, whether they involve work, consumption, health, travel, or liberty. At the organizational level, formal surveillance involves a constituency. This term is used broadly to refer to those with some rule-defined relationship or potential connection to the organization. This may involve formal membership or merely contact with the organization, as through renting a video or showing a passport at a border.
Organizations have varying degrees of internal and external surveillance. Erving Goffman has identified many kinds of employee or inmate monitoring, such as within “total institutions.” These offer examples of the internal constituency surveillance found in organizations. Here individuals “belong” to the organization in a double sense. First they belong as members. But they also in a sense are “belongings” of the organization. They are directly subject to its control in ways that non-members are not. Thus, there is a loose analogy to the ownership of property.
External constituency surveillance is present when those who are watched have some patterned contact with the organization, for example, as customers, patients, malefactors, or citizens subject to the laws of the state. Those observed do not “belong” to the organization the way that an employee or inmate does. Credit card companies and banks, for example, monitor client transactions and also seek potential clients by mining and combining databases. The control activities of a government agency charged with enforcing health and safety regulations is another example. In this case, the organization is responsible for seeing that categories of person subject to its rules are in compliance, even though they are not members of the organization. The same compliance function can be seen with nongovernmental organizations that audit or grant ratings, licenses, and certifications.
In the case of external non-constituency surveillance, organizations monitor their broader environment in watching other organizations and social trends. The rapidly growing field of business intelligence seeks information about competitors, social conditions, and trends that may affect an organization. One variant of this is industrial espionage. Organizational planning (whether by government or the private sector) also requires such external data, although this is usually treated in the aggregate instead of in personally identifiable form.
Personal surveillance, in which an individual watches another individual (whether for protection, strategic, or prurient reasons) apart from an organizational role, is another major form. It may involve role relationship surveillance as with family members (parents and children, a suspicious spouse) or friends looking out for and looking at each other (e.g., monitoring locations through use of a cell phone). Or it can involve non-role relationship surveillance such as in the free-floating activities of the voyeur whose watching is unconnected to a legitimate role.
With respect to the roles that are played, the surveillance agent (watcher/ observer/seeker) can be identified. The person about whom information is sought is a surveillance subject. All persons can play both roles, although hardly in the same form or degree. These roles may change depending on the context and over a per-son’s life cycle; the roles are also sometimes blurred and may overlap.
Within the surveillance agent category, the surveillance function may be central to the role, as with police, private investigators, spies, work supervisors, and investigative reporters. Or it may simply be a peripheral part of a broader role whose main goals are elsewhere. Illustrative of this are check-out clerks who are trained to look for shoplifters, or dentists who are encouraged or required to report suspected child abuse when seeing bruises on a child’s face.
A distinction rich with empirical and ethical implications is whether a situation involves those who are a party to the generation and collection of data (direct participants) or instead involves third parties. Third parties may legitimately obtain personal information through contracting with the surveillance agent (e.g., to carry out drug tests or to purchase consumer preference lists). Or information may be obtained because confidentiality has been violated by the agent, or because an outsider has obtained it illegitimately (e.g., wiretapping, hacking, corrupting those with the information).
The presence of third parties raises an important secondary-use issue: can data collected for one purpose be used without an individual’s permission for unrelated purposes? In Europe the answer generally is no. In the United States, where a much freer market in personal information exists, there are fewer restrictions. Large organizations warehouse and sell vast amounts of very personal information, without the consent and with no direct benefit to the subject.
Surveillance can also be analyzed with respect to whether it is nonreciprocal or reciprocal. Surveillance that is reciprocal may be asymmetrical or symmetrical with respect to means and goals. In a democratic society, citizens and government engage in reciprocal, but distinct forms of mutual surveillance. Citizens can watch government through Freedom of Information Act requests, open hearings and meetings, and conflict of interest and other disclosures required as a condition for running for office. However, citizens can not legally wiretap, carry out Fourth Amendment searches, or see others’ tax returns. In bounded settings such as a protest demonstration, there may be greater equivalence with respect to particular means; for example, police and demonstrators may videotape each other.
Agent-initiated surveillance, which is particularly characteristic of compliance checks such as an inspection of a truck or a boat, can be differentiated from subject-initiated surveillance such as submitting one’s transcript, undergoing osteoporosis screening, or applying for a job requiring an extensive background investigation. In these cases, the individual makes a claim or seeks help and essentially invites, or at least agrees to, scrutiny.
With agent-initiated surveillance, the goals of the organization are always intended to be served. Yet this need not necessarily conflict with the interests of the subject-the protection offered by school crossing guards and the efficient library service that depends on good circulation records benefit the organization and the subject. Public health and medical surveillance have multiple goals, protecting the community as well as the individual. Efficiently run companies provide jobs and services. Providing a limited amount of personal information on a warranty form and having a chip record usage of an appliance, such as a lawn mower or a car, may serve the interest of both consumers and businesses (e.g., being notified if the manufacturer finds a problem or offering proof of correct usage if the device fails).
Subject-initiated surveillance may reflect goals that serve the interests of the initiator, but they often overlap the goals of the surveilling organization. Examples include some protection services that have the capability to remotely monitor home and business interiors (video, audio, heat, gas, motion detection) or health systems for the protection of the elderly and ill (e.g., an alarm is sent if the refrigerator of a person living alone is not opened after 24 hours). Since these are forms of surveillance more likely to involve informed consent, these are less controversial than secretly generated agent surveillance. What is good for the organization may also be good for the individual, although that is not always the case and depends on the context.
II. Surveillance Processes
Rather than being static and fixed at one point in time, surveillance can be viewed as a fluid, ongoing process that involves interaction and strategic calculations over time. Part of the fascination with the study of surveillance lies in its dynamic nature, as groups in conflict relationships reciprocally and continuously adjust their tactics to each other. Here the moral ambivalence that infuses the topic as a result of conflicting values and social interests can also be seen.
The myth of surveillance involves creating and sustaining the belief through the mass media that a technique is omnipresent and omnipotent. When such literal watching (whether real or metaphorical) is not possible, there is an effort to create uncertainty about whether or not surveillance is present. This is presumed to be a deterrent. However, in a complex world of conflicting interests, unexpected developments, and technology that breaks, perfect knowledge and control are rarely possible. New conditions may appear and efforts to resist surveillance are common.
A number of behavioral techniques of neutralization—strategic moves by which subjects of surveillance seek to subvert the collection of personal information. Among these are direct refusal, discovery, avoidance, switching, distorting, coun-tersurveillance, cooperation, blocking and masking. Responses to drug testing illustrate most of these.
One type of response is refusal—just saying no or feigning an inability to offer the sample in spite of trying. Social systems often leak, and the date of a supposed random or surprise test or search can sometimes be inferred involving a discovery move. Employees may receive notice of a test. Such foreknowledge permits avoidance moves involving abstinence, the hiding or destroying of incriminating mate-rial, not going to work on that day, or leaving early because of illness. With switching, drug-free urine (whether purchased commercially in liquid or powdered form, or obtained from a friend) replaces the person’s own urine. Distorting responses (e.g., diuretics and commercially available detox products) manipulate the surveillance-collection process such that while offering technically valid results, invalid inferences are drawn. Given empathy and the multiplicity of actors and interests in complex organizations, various cooperative moves in which controllers aid those purportedly controlled may also be seen.
Counter surveillance involves an ironic turning of the tables in which the very technologies used to control others come to be used to advance the interests of those controlled. Thus, when facing a urine drug test, employees can first experiment at home, testing themselves with a variety of readily available products like those used in the official test. The wide availability of new tools such as covert audio and video recording offers the surveilled the chance to turn the usual stratification tables. The resulting data can be used defensively or to coerce those in positions of authority since there is the potential, for example, for the documentation of unwanted sexual advances and police abuse.
Another set of processes involves decisions about whether or not to surveil, and if so which technique to use, how to apply it, and to whom (e.g., to everyone, ran-domly, or selectively based on criteria). In addition, decisions need to be made about: whether or not the surveilled are to be informed of and have any say in the process or in how the data are used; who will have access to the data; what the degree of security will be; and, how long the data will be kept. The social scientist seeks to identify and explain these patterns and their prior correlates.
Such decisions have consequences. Subsequent developments can be contingent on the choices made. For example, some techniques that can be narrowly applied, such as the use of DNA for identification purposes, will require the creation of large databases against which a sample can be compared. A decision to watch everyone (categorical suspicion) avoids claims of discrimination in targeting, but it is more expensive and can lead to widespread feelings of privacy invasion.
Another aspect of process involves the path or “career” of surveillance events. In many, perhaps a majority, of cases, no action follows from surveillance because nothing of interest is discovered. The surveillance is intended to serve a scarecrow function or simply to generate a documentary record. Information may be saved until it is needed or a critical amount has been obtained. Or occasionally, too much is discovered to act on, or the surveilled can exert counter pressure to prevent action from being taken.
Yet techniques, too, have careers. The surveillance appetite can be insatiable and often shows a tendency to expand to include new goals, agents, subjects, and forms. Awareness of this requires asking of any new tactic, regardless of how benignly its means and ends are presented, where might it lead? The expansion of a technology introduced in a limited fashion can often be seen. Extensions beyond the initial use, whether reflecting surveillance creep or, in many cases, surveillance gallop, are common. There may be new uses for the data. For example, the Social Security number that Congress intended only to be used for tax purposes has become a de facto national ID number, and video cameras once restricted to prisons and high-security areas, are now found in offices, shopping malls, and homes.
New surveillance agents and subjects may appear. A common process is a progression from use on animals to prisoners, criminal suspects, noncitizens, the ill, children, and then throughout the society. Tactics developed by government for defense and law enforcement often spread to manufacturing and commercial uses, and then to uses in interpersonal relations by friends, family, and others. One example of this trend is the expansion of drug testing from the military to sensitive categories such as transportation workers to the workforce at large, and then even to parents who test their children. The patterning of the use of global positioning satellite data is equivalent. Yet expansion is only one path. A technique may be developed but not widely used; for example, this has been the case with handwriting analysis (graphology) in the United States, although not in France. Or if adopted, a tactic may diffuse slowly rather than rapidly throughout the social order. Télévision, for example, had been available since the late 1920s, decades before it came into widespread use.
Sometimes a rarely studied phenomenon of surveillance contraction occurs. Widely used tactics may come to be less used as a result of political controversy, the development of regulations, and unintended consequences, or as a result of the development of better tactics. For example, congressional legislation in 1988 severely restricted the use of the polygraph for employment purposes.
III. What Is to Be Done?
Social understanding and moral evaluation require attending to the varied contexts and goals of surveillance. The many settings, forms, and processes of surveillance preclude any easy explanations or conclusions. Two broad, con-trasting views of the new surveillance can be identified. A pessimistic Franken-steinian/Luddite view holds that surveillance technology is inhuman, destructive of liberty, and untrustworthy. In addition, since surveillance technology devel-ops out of an inequitable social context, it is seen as likely to reinforce the status quo.
A more optimistic view places great faith in the power of technology, which is seen to be neutral. More powerful surveillance tools are seen as necessary in today’s world, where efficiency is so valued and where there is a multiplicity of dangers and risks. Surveillance is a sword with multiple edges. The area is fascinating precisely because there are no easy scientific or moral answers.
There are value conflicts and ironically conflicting needs and consequences, all of which make it difficult to take a broad and consistent position in favor of, or against, expanding or restricting surveillance. For example, society values both the individual and the community. Members of society want both liberty and order. Individuals seek privacy and often anonymity but also know that secrecy can hide dastardly deeds, and visibility can bring accountability. Although individuals value freedom of expression, too much visibility may inhibit experimentation, creativity, and risk taking.
As with any value-conflicted and varied-consequence behavior, particularly those behaviors that involve conflicting rights and needs, it is essential to keep the tensions ever in mind and to avoid complacency. Occasionally, when wending through competing values, the absolutist, no compromise, don’t cross this personal line or always cross it standard is appropriate. But more often compromise—if rarely a simplistic perfect balance—is required. When privacy and civil liberties are negatively affected, it is vital to acknowledge, rather than to deny this impact, as is so often the case. Such honesty can make for more informed decisions and also serve an educational function.
Surveillance practices are shaped by manners, organizational policies, laws, and by available technologies and counter technologies. These draw on a number of background value principles and tacit assumptions about the empirical world that need to be analyzed. Whatever action is taken, there are likely costs, gains, and trade-offs. At best we can hope to find a compass rather than a map, and a moving equilibrium rather than a fixed point for decisionmaking. The following questions (Marx, 2005) can help.
Questions for Judgment and Policy Regarding Surveillance
1. Goals: Have the goals been clearly stated, justified, and prioritized?
2. Accountable, public, and participatory policy development: Has the decision to apply the surveillance technique been developed through an open process and, if appropriate, with the participation of those to be surveilled? This involves a transparency principle.
3. Law and ethics: Are the means and ends not only legal but also ethical?
4. Opening doors: Has adequate thought been given to precedent creation and long-term consequences?
5. Golden rule: Would the watcher be comfortable with being the subject rather than the agent of surveillance if the situation were reversed?
6. Informed consent: Are participants fully apprised of the system’s presence and the conditions under which it operates? Is consent genuine—beyond deception or unreasonable seduction—and can participation be refused without dire consequences for the person?
7. Truth in use: Where personal and private information are involved, does a principle of unitary usage apply, in which information collected for one purpose is not used for another? Are the announced goals the real goals?
8. Means-ends relationships: Are the means clearly related to the end sought and proportional in costs and benefits to the goals?
9. Can science succeed? Can a strong empirical and logical case be made that a means will in fact have the broad positive consequences its advocates claim? Does the system or technique really work?
10. Competent application: Even if it works in theory, does the system (or operative) using it apply it as intended?
11. Human review: Are automated results with significant implications for life chances subject to human review before action is taken?
12. Minimization: If risks and harm are associated with the tactic, is the tactic applied to minimize these, displaying only the degree of intrusiveness and invasiveness that is absolutely necessary?
13. Alternatives: Using a variety of measures, not just financial ones, are alternative solutions available that would meet the same ends with lesser costs and greater benefits?
14. Inaction as action: Has consideration been given to the “sometimes it is better to do nothing” principle?
15. Periodic review: Are regular efforts made to test the system’s vulnerability, effective-ness, and fairness and to review policies?
16. Discovery and rectification of mistakes, errors, and abuses: Are there clear means for identifying and fixing these and, in the case of abuse, applying sanctions?
17. Right of inspection: Can individuals access and challenge their own records?
18. Reversibility: If evidence suggests that the costs outweigh the benefits, how easily can the surveillance be stopped? What is the extent of capital expenditures and available alternatives?
19. Unintended consequences: Has adequate consideration been given to undesirable consequences, including possible harm to watchers, the watched, and third parties? Can harm be easily discovered and compensated for?
20. Data protection and security: Can surveillants protect the information they collect? Do they follow standard data protection and information rights as expressed in the Code of Fair Information Protection Practices and the expanded European Data Protection Directive?
A central point of much social and legal analysis is to call attention to the con-textual nature of behavior. Certainly these questions and the principles implied in them are not of equal weight, and their applicability will vary across time peri-ods, depending on need and perceptions of crisis, across contexts (e.g., public order, health and welfare, criminal and national security, commercial, private individuals, families, and the defenseless and dependent), and particular situations within these. Yet common sense and common decency argue for consider-ing them.
