A republican account of the value of privacy

There is a substantial literature on privacy and a steadily expanding body of contemporary writing on republican political theory. However, republican writers have paid little attention to the way in which privacy (or a right to privacy) serves republican aims. Indeed, there seems to be a widespread and faulty perception that privacy serves individual interests in a way that undermines republican ideas of civic duty and promotion of the collective good, by allowing individuals to avoid their collective responsibilities and by concealing privilege and power that cannot be justified on republican grounds. ¹ The position defended in this article is that privacy is an important republican value; a value that serves a central ideal of modern republican theory – freedom as non-domination. While the privacy literature encompasses various liberal perspectives, it lacks a well-developed republican account of the value of privacy. The aim here is to provide such an account. In the course of doing so, I will show that a republican account can provide us with an explanation as to why loss of privacy is harmful in circumstances that will prove to be problematic cases for liberals – problematic in the sense that liberals will agree that privacy is valuable, and that loss of privacy is harmful, but will find it difficult to explain why this is the case.

Republicanism is an egalitarian theory of self-government, the aim of which is to maximize conditions of freedom so that individuals are able to pursue self-determined ends. These aims, which are broadly similar to those envisaged by liberals, reflect a common heritage. But while it is possible to trace the roots of modern liberal theory back to classical republican ideas, those who have contributed to the modern revival of interest in republican thinking adopt differing positions on its relationship with liberalism. Some set republicanism against liberal theory, claiming that republican ideals not only differ fundamentally from those that are central to liberal theory but are also superior to and ought to displace liberal ways of thinking. ² Others have drawn attention to classical republicanism as the common origin of both modern liberal and republican theories and deny that there is a sharp distinction. ³ But while liberalism might be viewed as a derivation of classic republicanism, most modern republican theorists agree that liberalism has lost its way; that it has developed in ways that, when compared with republicanism, reveal it to be ‘impoverished and incoherent’. ⁴ The main objects of the republican critique of liberal doctrine are the conception of freedom as absence of inference, and liberalism's failure to acknowledge that individuals can only be assured of their freedom through active participation in political decision making. While republicanism is sometimes caricatured as a doctrine that favours collective over individual interests, contemporary republican theorists point out that modern republican theory ‘fully endorses the moral individualism and ethical pluralism of modern society, and does not deny the existence of individual rights’. ⁵

Is it possible for republican citizens to have a right to privacy? If so, what are its justificatory grounds? In what follows, it will be suggested that the value of privacy for republicans lies in its capacity to shield individuals from the threat of domination. A consequence of loss of one's privacy is that others may acquire dominating power – the capacity to interfere in one's decisions on an arbitrary basis. There seems to be much anxiety about the extent to which governments and large corporations collect and retain information about us. For most of us, the collection and retention of such information does not result in any obvious interference in our affairs. How then can we account for this anxiety, and do we have good reason to be anxious? If we suffer no interference as a result of loss of privacy, is our freedom really diminished by the loss? What if, in addition to suffering no interference as a result of a loss of privacy, we are unaware that there has been some interference with our privacy? These are questions to which few satisfactory answers seem to have been provided. I will suggest that at the root of our anxieties about loss of privacy, is concern about what republicans will recognize as domination. While liberals can explain how a loss of privacy that does not lead to any interference might affect the individual's autonomy, where she is unaware of the loss, they will have difficulty in explaining the value of privacy. However, the republican conception of freedom as non-domination provides us with the foundation of a coherent account of the value of privacy – one that allows us to say that the subject may, in fact, suffer a diminution of her freedom in such circumstances. Much of the republican account that is provided here could be accommodated by a liberal account of privacy, as one might expect of theories that have a common ancestry, but they are not co-extensive. There are circumstances in which the republican account provides us with answers as to why privacy is valuable that are not provided by existing liberal accounts.

The first section of the article describes the republican conception of freedom as non-domination and the various broad forms that dominating power might take. In the second section, the focus shifts to the means of securing freedom (conditions of non-domination). Republicans believe that members of a polity can only be assured of such conditions if they participate in the political decision-making processes that will determine the content of the laws to which they are subject, and the nature and scope of the rights that they will hold. The third section explores the role of privacy in securing conditions of freedom in the face of the threat of various forms of dominating control. In the final section, the role that privacy performs in facilitating effective participation in political decision-making processes is considered. Here it will be suggested that privacy is a pre-requisite for such participation. It facilitates restricted forms of deliberation that are an essential prelude to the broad and inclusive public forms of deliberation through which the norms that will determine the nature of the relationship the citizens have with one another in a republican democracy, and the relationship that citizens have with the state. It will also be argued that effective participation in political decision making will be dependent on citizens having a more general guarantee of privacy – a guarantee that secures the conditions of non-domination that are required for personal autonomy. The individual who is unable to develop an autonomous way of life generally is likely to be incapable of exercising the kind of political autonomy that is required, if the republican ideal of self-government is to be realized. It will be suggested that personal and political autonomy are mutually sustaining, and that privacy is a pre-requisite for both.

Freedom as non-domination

As noted in the introduction, republicanism aims to secure conditions of freedom for the citizens of a self-governing polity. In some versions of republicanism, participation in political life is seen as an end in itself – that individuals are essentially social and political beings and realization of this can only be achieved through participation in political life. ⁶ Others think that it is instrumentally valuable – that it is only through participation in political processes that individuals can be assured of their freedom. ⁷ It is to this version of republicanism that reference is made in asking whether, and if so how, privacy might serve republican ends, and references to ‘republicans’ will be to those who subscribe to this version of republican theory.

Republicans claim that the way in which they conceive freedom constitutes an important point of distinction between republican and liberal theories. Privacy is associated with the possibility of freedom and so if the republican conception of freedom is distinct from its liberal counterpart, as republicans claim, it raises the possibility of an account of the value of privacy that is distinctive insofar as it can in some respects be distinguished from a liberal account.

Liberals say that conditions of negative freedom exist insofar as there is an absence of interference, so that a subject can be said to be free to the extent that her choices or preferences are not overridden by means of another's intentional obstruction, coercion, deception or manipulation. ⁸ For republicans, this conception of freedom fails to encompass the full range of circumstances in which it can properly be said that there is some diminution in individual freedom. As we will see, while republicans say that only some forms of interference constitute a loss of freedom, they also claim that an individual can suffer such a loss in the absence of any interference. If individuals are to enjoy conditions that are a pre-requisite for the successful pursuit of self-determined ends, we need a more expansive conception of freedom, one that is captured in the idea of non-domination.

Perhaps the best-known definition of this concept is offered by Phillip Pettit, who suggests that someone dominates another to the extent that the person has the capacity to interfere, on an arbitrary basis, in choices that the other is in a position to make. ⁹ Interference, he explains, will be arbitrary if it fails to track the avowable interests of the subject – that is to say, interference is ‘chosen or rejected without reference to the interests, or the opinions of those affected’. ¹⁰ An act of interference is to be considered arbitrary, not by reference to the particular consequences to which it gives rise but where there is a lack of control over the interference – where the interference is unchecked. ¹¹

In more recent writing, Pettit provides a refined definition of domination in which the idea of control replaces the concept of arbitrariness; ‘Someone, A, will be dominated in a certain choice by another agent or agency, B, to the extent that B has a power of interfering in the choice that is not itself controlled by A’. ¹² The power to interfere will be ‘controlled by A’ only if it is exercised ‘in a direction or according to a pattern that A has the influence to determine’. ¹³ Where reference is made in what follows to ‘arbitrary interference', it is to interference that is not controlled in this way. One can experience interference then, without being subject to dominating power. To take a simple example, Smith might think that it will help him in his attempt to give up smoking to issue an instruction to all of his friends, that if they see him with a cigarette in his mouth they should, without hesitation, remove and extinguish it. If Brown does just this, he interferes with Smith's choice to light up, but does so in accordance with directions that Smith has issued. The interference is, therefore, an exercise not of Brown's preferences as to whether Smith should smoke but of Smith's desire that his friends assist him in his attempt to give up smoking. In the absence of any retraction of the instruction that Brown has reason to take seriously, his repeated interventions to prevent Smith from lighting up does not constitute dominating control – the interference is in accordance with Smith's direction.

Whether power to interfere constitutes dominating power will depend on whether it can be said that the person who is subject to it exercises control over the way in which it will be used. The concept of control, like other broad concepts, is ambiguous. Various questions arise as to what constitutes control and what degree of control is required before we can say that although others have power to interfere in our choices, we are not subject to dominating control. It will not be possible to pursue these issues at length here, but one point should be made before we move on. Save in those circumstances in which a subject lacks the capacity to engage in rational decision making, paternalistic interference is inconsistent with the idea of freedom as non-domination. A person might interfere in your choices where he or she believes that your interests will not be served by the choice that you would make if you were to be given free rein. Pettit suggests that such interference involves the interferer substituting his own interpretation of what is in your best interests, and discounting yours as inferior. It is interference that gives effect to his will, not yours, and you are therefore subjugated to dominating power. ¹⁴

I can interfere paternalistically not only with decisions that you are in a position to make regarding the ends that you will pursue but also with your decisions regarding the means of achieving ends that you have chosen for yourself. Returning to Smith's efforts to give up smoking, let us suppose that Brown having been given a direction by Smith that if sees him with a cigarette he should remove and extinguish it, decides instead that he will replace Smith's tobacco filled cigarettes with relatively harmless nicotine-free substitutes. Suppose also that Smith will not be able to tell the difference between the real and the substitute cigarettes. Does this amount to a form of dominating control? The answer must be that it does. Interference in an agent's decision making in order to secure some end that the interferer believes to be in the agent's best interests rejects the agent's interpretation of what is in his best interests, and replaces it with the interferer's interpretation. Similarly, interference within an agent's decisions as to the manner in which some end ought to be secured (in a way that is not controlled by the agent) replaces the reasons, principles and values that underpin the agent's decision about appropriate means, with reasons, principles and values that guide the alternative course of action.

Pettit points out that interference in an agent's choices might take various forms. First, options that may have been open to the agent might simply be withdrawn by someone who has the power to remove an option from those available, so that it is no longer possible for the agent to choose it. Of course, it may not be possible to remove certain options. The interferer may not possess the power, or control over forces, that would be required to remove a particular option. The option may be of a kind that cannot be removed – it may be one that will always be open to the agent. But although removal of an option might be impossible, it may still be possible to interfere with an agent's choices by making the option considerably less attractive, for example, by attaching sanctions to it, so that the agent would incur considerable costs were he to choose it. In such circumstances, the original unencumbered option is replaced with a far less attractive alternative.

The final mode of interference is manipulation of an agent's decision making through the misrepresentation of options. ¹⁵ Successful manipulation results in the agent being unable to make a choice on a proper understanding of the choices that are available. She may be ‘nudged’ towards certain options and away from others. An interferer might attempt to influence an agent's decision making by overloading her with so much information that any strategic vision is clouded, blinding her as to the nature of certain options that are available. Rewards and incentives may be offered, and attempts may be made to induce feelings of guilt for not choosing an option that the manipulator would prefer the agent to choose.

It should be noted that each of these modes of interference – the removal, replacement or misrepresentation of options – can be pursued in the presence or the absence of awareness on the part of the victim of the interference. Its general effect as a form of dominating control, however, is likely to be amplified where the victim is aware.

This outline will do for the time being. I will provide some concrete examples of the forms of interference described above in the course of explaining, in the next section, how conditions of privacy might frustrate attempts to interfere in an agent's choices. For present purposes, we should note that republicans are concerned about interference, but not interference per se. Concern is reserved for others' capacity to interfere in an agent's choices on an arbitrary basis. The individual who suffers such interference is at the mercy of the agent or agency that has power to interfere. But while such interference will always constitute domination – to a greater or lesser extent, depending on the nature of the interference – a person need not interfere with another's choices in order to exercise dominating control. If an agent or agency has the power to interfere arbitrarily in an individual's choices, freedom is diminished even if the power is never exercised.

For republicans, the paradigmatic case of domination is servitude. A person who is enslaved might have the good fortune to fall into the hands of the most benevolent of owners, a master who gives him a free rein. However, republicans point out that even though such a person may find that he generally gets by without being subjected to arbitrary interference, this state of affairs is at all times contingent on the good will of his master. Although the slave is seemingly free to make his own decisions, those decisions will be taken in light of the knowledge that anything he chooses to do may be thwarted if it does not meet with his master's approval. Although there is no active interference in the slave's affairs, republicans would not say that he enjoys freedom in any meaningful sense of what it means to be free. The republican idea of freedom is perhaps best summed up by Duncan Ivison:

I am not free simply because no one actually interferes with me … [T]he bare potential for interference threatens my liberty, not only the likelihood or probability of its exercise, and that means that as long as it is counterfactually the case that someone could arbitrarily interfere with me, I am unfree. So it is not merely the prospect or probability of arbitrary interference that matters but the bare potential. ¹⁶

To be exposed to dominating power is to live with the high degree of uncertainty that comes with not knowing whether or when one will be subject to interference. Republicans claim that not only is this state of affairs likely to engender considerable anxiety and a sense of dispiritedness ¹⁷ but it also makes planning much more difficult than it would be under corresponding conditions of non-domination. Where another has the power to interfere in an agent's choices according to his whims, preferences or desires, the agent will be required to engage in strategic deference, or to anticipate interference and take evasive action in attempt to ensure that he provides the other with no reason to exercise the power he possesses. ¹⁸ The development of successful strategies of deference and evasion, and the planning that this entails imposes significant costs, requiring the diversion of resources that could otherwise be deployed in determining and pursuing ends that the individual considers valuable. But, to be clear, an agent's awareness that another has the power to interfere in his decision making does not constitute a necessary condition of domination. If another has power to step in at any point to prevent the agent giving effect to his preferences, he is subject to dominating control, whether or not he is aware that the other is able to do this. Although from the agent's perspective it might appear that he is free to act as he sees fit, his freedom is contingent on the way in which the other chooses to exercise the power to interfere.

Securing freedom

The way in which freedom is conceived is said by republicans to mark a significant difference in republican and liberal thinking. A second ground on which it is claimed that the two can be distinguished is the republican belief that an individual can only be assured of her freedom through participation in politics. Recall that, according to Pettit, a person will be dominated in a certain choice by another agent or agency to the extent that the agent or agency possess a power to interfere that is not itself controlled by the person. Control exists where the power is exercised in accordance with a direction issued by the person who is subject to it, or according to a pattern that he or she has the influence to determine. This idea was illustrated earlier using the rather trivial example of a smoker who issues a directive to his friends to extinguish any cigarette that they see him smoking. But suppose we think about interference not by well-meaning friends but by a more remote and powerful agency. How is the individual to control the power that the state has to interfere in the decisions that he is in a position to make? Some say that he need do little, so long any measures that might be adopted track common interests. The participatory duty of the citizen ex ante extends no further than the election of representatives who will be expected to ensure that the interests of those who elect them are given due consideration and weight in political decisions. If any measure is adopted that fails to track the interests that they share in common with others, citizens must be prepared to step in and contest it. Those who suggest that citizens' ex ante participatory duties are in this way relatively undemanding, claim that the need to secure electoral support and to minimize the likelihood of decisions being challenged mean that representatives will ensure that their decisions take proper account of the interests of those they represent. ¹⁹

The notion of the passive but vigilant citizen is, however, at odds with the classical republican idea of self-government. As Robin Celikates points out,

whether a (basic) norm is compatible with the freedom of those subject to it depends on whether the latter have had (the chance to have) a say in the framing of the norm (and continue to have a say in its continuing application), not on whether they would or could – hypothetically or ex post – give their consent to it. ²⁰

To entrust others with this task, he suggests, is to ‘outsource’ self-determination. The idea of self-government requires citizens to understand the norms to which they are subject as an expression of their own political practice. ²¹ The members of a republican democracy can only realize the ideal of self-government, and be sure that they will enjoy conditions of non-domination, through active participation in the decision-making processes that generate – determine the nature and extent of – the norms that will regulate their conduct. Participation may take various forms, of course – from contributions to discussion on the development of political policies generally, ²² to ex ante contestation of concrete proposals and submission of counter proposals as part of a formal consultation process, and the empanelling of ad hoc citizens’ assemblies to generate legislative proposals on issues of fundamental importance. ²³ But with membership of a self-governing polity comes the expectation that citizens will, when necessary, set aside private interests and pursue the public good of ensuring that the laws to which individuals are subjected track their common interests.

The expectation that citizens will hold these pre-dispositions or commitments comprises a central idea of the republican concept of civic virtue, and in this respect republican civic virtue is more demanding than its liberal counterpart. In the liberal state, citizens' involvement in political life, and support of liberal institutions, will be considered virtuous. However, the inculcation of a commitment to these institutions, the adoption of measures that are intended to instill in citizens a disposition to engage in political decision-making processes and to acquire the attributes required if their participation is to be effective are likely to be considered by liberals to be a threat to state neutrality. Republicans have pointed out, however, that in abandoning the notion of state neutrality, they seek to isolate and promote only one strand in citizens' conceptions of the good; the good of not having to guard against arbitrary interference in one's affairs. ²⁴ Because conditions of non-domination enable citizens to identify and pursue their own conceptions of the good, the expectation that citizens will act in ways that promote and secure non-domination should not be considered inherently oppressive.

The decision-making processes in which the citizens of a republican democracy have a civic duty to participate ought to be deliberative. Power conferred on officials must serve some interest that members of the polity share in common if it is not to constitute dominating power. Establishing just what these interests are, and that the conferral of power will, in fact, serve them requires a form of deliberation that ensures that any decision is taken in light of a broad range of views. It is particularly important that the views of minority groups and sections of the community that might be disproportionately affected by the proposals are represented. The process should facilitate contestation. There must be adequate opportunities for opponents to challenge proposals and to present counterproposals. Interventions must be taken seriously and should be met by meaningful responses. Throughout the process, all parties should reflect on alternative points of view, be willing to revise or abandon their positions in light of new information and insights, as well as being prepared to explain by reference to the grounds on which the position they advance is challenged, why it should stand. In sum, well-functioning deliberative processes will be those that provide citizens with an adequate opportunity to influence the content of legislative proposals that provide officials with power to interfere in the choices that they – the members of a self-governing polity – are in a position to make. Even if the views held by some who participate in these processes do not prevail, their active involvement might still enable them to exercise a form of control. The contester may not get his way but he can at least be assured that any measure that is adopted is supported by public reason – that it does, in fact serve common rather than sectional interests.

This relatively brief outline of two important strands of republican theory – freedom as non-domination and the importance of the polity's participation in political decision making in securing conditions of freedom – provide us with enough of a framework to start thinking about why republicans might value privacy.

The value of privacy for republicans

Let us, for present purposes, take privacy to be the condition in which the individual is (or a group of individuals are) separated from others, so that an agent has privacy where others do not have physical access to him, or access to information about him. Privacy is then, associated with the idea of negative freedom – those who have privacy do not suffer certain forms of intrusion. We have noted that republicans claim that the republican conception of freedom is distinctive, that it constitutes a necessary extension of the liberal conception of freedom as non-interference. I suggested earlier, that in light of this, a republican account of the value of privacy might be similarly distinctive, and it is to this possibility that we now turn.

It should first be said that republicans might value privacy for various reasons. Some have suggested that privacy is inherently valuable, that it is part of human nature to seek a degree of separation from others in some circumstances, and that such separation is essential to human flourishing, and there is no reason to think that republicans would reject this claim. ²⁵ Republicans might also be expected to accept that privacy is a pre-requisite for autonomous agency and that, more broadly, respect for individual privacy can be conceived of an aspect of the principle of respect for human dignity. However, the focus in what follows will be on republican concern with freedom as non-domination. I will consider how privacy might protect an individual from the various forms of interference identified by Pettit: the replacement, removal and misrepresentation of options. This part of the republican account of privacy does not differ substantially from standard liberal accounts. I will also explore the relationship between privacy and that aspect of freedom that republicans argue is a necessary extension of the liberal conception of freedom. This is the loss of freedom that republicans say can occur in the absence of interference – the subjugation of a person to dominating power by virtue of another merely having the capacity to interfere on an arbitrary basis. It is here that a distinctive republican account of the value of privacy emerges. Where an agent is aware that he has suffered a loss of privacy but has been subject to no interference, it is possible for liberals to construct an account of the value of privacy around a positive conception of freedom. This account cannot, however, accommodate circumstances in which the agent is unaware that he has suffered such a loss – that he has been the subject of covert surveillance, for example, or that others have acquired information about him without his knowledge. In contrast, a republican account of privacy grounded on the idea of freedom as non-domination can explain a loss of privacy where there is no subsequent interference in terms of negative freedom. Further, this explanation covers both circumstances in which the subject is aware of the loss, and those in which he is not.

As we have already noted, exercise of the power to interfere will not be arbitrary if the person who is subject to it is able to exercise some control over it, and active participation in political decision-making processes provides citizens with an opportunity to control the use of power. I will describe the role that privacy plays in facilitating such participation. It will be suggested that republicans ought to consider privacy to be a reflexive, common good; that the right to privacy provides citizens with conditions of individual freedom that enable them to lead self-directed lives, but it is also a pre-requisite for the political autonomy that will be required for effective participation in processes through which a guarantee of privacy (and other basic liberties) can be secured.

Privacy, domination and interference

If a republican account of the value of privacy was to be restricted to consideration of the various circumstances in which the threat of domination came in the form of another's interference with an agent's choices, it would look rather like a standard liberal account. Liberals value privacy because they recognize it to be a pre-requisite for the development of autonomous ways of life. While personal autonomy is always in the foreground of liberal thinking, there are few references to it in republican writing. But this should not be taken as an indication that republicans are unconcerned about autonomy. Pettit explains that autonomy is a far richer ideal than that of freedom as non-domination, but claims that the republican focus on non-domination is warranted on the grounds that ‘it is bound to be easier for people to achieve autonomy once they are assured of not being dominated by others’. ²⁶ The republican pre-occupation with domination is premised on the belief that if people are assured that they will not be exposed to dominating control, they can be ‘trusted to look after their own autonomy’. ²⁷ But as Richard Dagger points out, when republicans attempt to explain the harm that results from subjugation to dominating power, they tend to do so by describing its adverse effects on an individual's capacity for autonomy. The person who suffers domination, he suggests, suffers ‘the wrong of not being respected as someone with a right to live, think, and speak for himself’ ²⁸ – an implicit appeal to autonomy. To be the subject of domination is to have questions as to what you can, may, or will do taken out of one's hands; such issues are ‘up to someone else’. ²⁹ If we think about Pettit's account of domination in the form of arbitrary interference in a person's choices, the relationship between autonomy and domination is evident. Arbitrary interference in an agent's choices may seriously impede an agent's attempts to determine for himself the kind of life that he ought to lead – to pursue self-determined ends. An individual might lack the capacity for autonomy in the absence of domination, but subjugation to dominating power will always diminish an individual's capacity for autonomy. Republicans will value privacy for much the same reasons liberals insofar as it protects a person from arbitrary interference in an individual's choices – in the form of removal, replacement or misrepresentation of options.

Let us start by considering how privacy – or a right to privacy – might ensure that others are unable to replace certain options – to substitute an option that you might be inclined to choose, with a less attractive alternative, so that you are more inclined to choose some other option that I would prefer you to choose. One broad way in I might achieve this (assuming I have the power to do so), is to attach a sanction to an option that I cannot remove from the range of options that are available to you. If the sanction is sufficiently onerous, you are likely to forego the option to which it is attached and choose an option I would prefer you to choose. Of course, my only preference might be that you do not choose the option to which I have now attached a sanction. But how might privacy shield a person from this kind of interference. Well, decisional privacy provides individuals with freedom to make fundamental decisions about the way that they live their lives – to choose their affiliations, to choose those with whom they wish to form intimate relationships and so on. It affords protection against attempts to ensure that those decisions conform to the preferences – moral or otherwise – of those who disapprove of the choices that the agent might make, if given free rein. Take, for example, section 337A of the Singapore Penal Code. This provision makes it an offence for any male ‘to commit any act of gross indecency with another male person’, whether the act takes place in a public or a private place. If convicted a person may be imprisoned for up to two years. Now, the state of Singapore cannot possibly remove the various opportunities that a man might have to engage in sexual activity with another man, so that it could be said that such a course of action is no longer de facto open to him. But it may be able to coerce him in a way that makes it less likely that he will take advantage of the opportunities that are available to him. It can do this by attaching a criminal sanction to such conduct. Criminalization replaces one option – the freedom to engage in sexual activity without any prospect of punishment by the state, with another – the option of engaging in such activity attended by the prospect of up to two years in prison if discovered. We might think the rational agent, if he perceives there to be a significant probability of being discovered, would be less inclined to engage in conduct that might be considered an act of ‘gross indecency’, than would be the case if that option was sanction-free. If this assumption is reliable, the failure to respect decisional privacy results in an interference in fundamental choices about the way in which the agent wishes to lead his life, and if this interference is arbitrary – that is to say, the agent has no opportunity to exercise control over the power – the interference constitutes a form of domination.

Moving on, examples of circumstances in which loss of privacy leads to the removal of options are not difficult to come by. It came to light recently in the United Kingdom that a group of large construction companies had established a database of 3200 ‘blacklisted’ workers – those who were perceived to be ‘troublemakers’, an enterprise that seems to have benefitted from the collusion of the police who were said to have been supplying information on various workers. ³⁰ The effect of the disclosure of information held on the database to potential employers would, no doubt, have resulted in foreclosure of any employment prospects in the construction industry – the removal of options that would have been available but for the violation of the subjects' informational privacy. Liberals might be expected to explain the harm caused by the disclosure by referring to its effects on the autonomy of the subject of the disclosure. The setback to employment prospects may frustrate career ambitions, or at a more base level, deprive the agent of an opportunity to use her skills and experience to generate the financial resources that would be required to realize the kind of life that she envisages for herself. Either way, the loss of privacy is likely to have a significant and adverse effect on the subject's autonomy. Republicans, on the other hand, are likely to say that the disclosure of information and its use by potential employers in recruitment decisions are forms of domination. If the existence of a database is not common knowledge – it is a secret database – members of the polity will have had no say in whether or not commercial entities ought to be permitted to establish such databases, and if they are, the terms on which they ought to be operated. In the absence of an opportunity to exercise control in this way, those who administer the database, and those who have access to it, acquire dominating control – a power to remove the employment options of those who are chosen for inclusion on the database.

Finally, privacy may protect an agent from the possibility that others will interfere with his choices by misrepresenting options that are open to him. Pettit, it will be recalled, suggests that misrepresentation of options may enable an interferer to manipulate the decision maker in a way that is likely to result in him choosing as the interfere wishes him to choose. He suggests that an interferer might be able to manipulate an agent's decision making by ‘snowing them with so much information that they are putty in [the interferer's] hands’. ³¹ Successful manipulation, he explains, ‘will affect the exercise of your cognitive capacity to choose between certain options even if it leaves your objective capacity in place. By means of manipulation I may succeed in getting you to choose as I wish’. ³²

Neil Richards provides an example that illustrates how loss of privacy facilitates this kind of manipulation. ³³ Surveillance, he explains, provides corporations with a large amount of information about customers, and this can be used to persuade them to buy products or services more often or to buy a wider range of goods or services. The supermarket chain, Target, for example, retains records of what its customers purchase. It uses that information as the basis for an inference as to which customers are likely to be pregnant – the purchase of unscented lotion and magnesium supplements, it seems, warrants such an inference. Once a customer is marked as someone who might be pregnant they can be subjected to ‘targeted marketing’. This involves flooding them with information (targeted advertisements) relating to products that they offer and which expectant mothers might purchase. This is done in the belief that if such customers are targeted in this way with sufficient frequency, they are likely to succumb to the suggestion that they ought to purchase the advertised items, and that they are likely to keep returning to the supermarket for this and other products for several more years. ³⁴ Now, some might consider this kind of marketing to be beneficial insofar as it saves those who are targeted the trouble of looking for products that they might need. Others might think it a distracting imposition. Dominating power can be measured in degrees, and while this kind of targeted marketing pales in comparison to the creation of secret databases that lead to people being denied employment, to the extent that those who are targeted have no say in the retention and use of personal information, it constitutes a form – albeit a minor form – of dominating control.

In relation to each of the examples provided above, republicans will say that loss of privacy could lead to interference, and that if the person who suffers the loss has no control or influence over the exercise of the power to interfere, it constitutes dominating control (and that consequently, the person suffers a diminution of freedom). Republicans will have no hesitation in claiming that privacy is valuable in these situations because it prevents others from acquiring dominating power. If pressed to explain why domination is harmful, they would no doubt offer an explanation that liberals would recognize as a claim that privacy is a pre-requisite for personal autonomy.

There is, then, nothing in the discussion of privacy so far, that suggests a republican account of the value of privacy is very different to its liberal counterpart. The grounds on which it can be distinguished come in the sections that follow, the first of which concerns the value of privacy in preventing domination in the absence of interference.

Privacy and domination in the absence of interference

The discussion above concerned the way in which privacy might protect a person from the threat of arbitrary interference in his choices. Two of the examples that were provided involved the use of information collected and retained on databases. But what about the mere possession of information about an individual? Suppose someone were to read my personal diary without my consent or without my knowledge. Most people would agree that I suffer a loss of privacy. Those who believe privacy to be a condition of inaccessibility per se might say that my consent is irrelevant to the question of whether I suffer a loss of privacy – the mere fact that another acquires information about me means that I do, in fact, suffer such a loss.

Suppose that having read my diary, the reader is now in possession of information that I would find embarrassing or shameful were others to become aware of it. Let us further suppose that the reader discloses it to another person, and in the act of disclosing it, it is overheard by a third person. Each now has knowledge of the embarrassing/shameful facts recorded in my diary. Presumably, we would all say that the person who reads my diary interferes with my privacy. We would probably be willing to say that the acquisition of this information both by the person to whom the reader discloses it and by the person who overhears the disclosure are further losses of privacy. But how do we explain the harm that I suffer as a result of others learning what I rather they did not know about me? We can explain why their use of that information in a way that interferes with the choices that I might make is harmful – but what is the harm to which mere possession of that information give rise?

It is not at all obvious if we take freedom to be an absence of interference – or even if we say that that freedom is diminished where it is probable that another will interfere ³⁵ – how the disclosure to the second person constitutes a loss of freedom. It is even less apparent how the acquisition of information by the person who simply overhears the disclosure constitutes any interference in my affairs or diminishes my freedom. That the person might use the information to interfere in my decision making is no more than a bare possibility – a threat, or a risk, that I will suffer some interference. I know nothing of their intentions. They may never use the information, but I take no comfort in this possibility. It will not ease my anxiety about the loss of privacy. If I am aware of the loss – that I know or realize that my diary has been read – then liberals can explain the harm that results from the loss of privacy in terms of positive freedom. An agent can be described as free not only to the extent that an adequate range of options are open to her but also to the extent that she has the capacity to make self-determined (authentic) choices in relation to those options. She might suffer a diminution in (negative) freedom if others interfere to close off certain options. But the degree of (positive) freedom that she enjoys may be diminished if her capacity for authentic decision making is undermined. Beate Rössler suggests informational privacy is valuable to people because it is an ‘intrinsic part of their self-understanding as autonomous individuals… to have control over their self-presentation… how they want to present or stage themselves, and how they want to be seen’. ³⁶ If an individual is aware that others know what she would prefer them not to know, there will be involuntary shift in her perspective, from the first to the third person. This shift prevents her from acting in an authentic, self-determined way. Her conduct will be conditioned by her awareness of what others know – it will prevent her from doing what she wishes without consideration for others' perspectives. Crucially, it may lead her to act in ways that are less likely to attract censorship and disapprobation; to adopt behaviour that she does not want to adopt. In light of this, Rössler suggests that loss of privacy in such circumstances is closely connected with the possibility of individual freedom. ³⁷ Here, the ideas of positive freedom and autonomy converge, so that we can say that the adverse effect on the diarist's autonomous decision making of others knowing the shameful or embarrassing events recorded in his diary diminishes his freedom to act in an authentic or self-regarding way. However, while the idea of positive freedom might provide the foundation for a plausible account of the value of privacy in cases in which the diarist is aware that he has suffered a loss of privacy, its reliance on the effect of the loss of privacy on the autonomy of the subject means that it cannot explain the value of privacy in cases where the subject is unaware that he has suffered a loss of privacy. Nor can the harmful effects of the loss be explained by reference to the conception of negative freedom as an absence of interference. If we believe that privacy is valuable in such circumstances – that others should not be permitted to subject us to covert surveillance or to covertly acquire and retain information about us – we need to look elsewhere for the basis of an explanatory account.

It seems to me that the republican idea of domination provides a concept that enables us to speak coherently about the relationship between privacy and negative freedom in circumstances in which a loss of privacy is not followed by any interference, and to do so whether or not the person who suffers the loss is aware of the loss. It was noted in the first section that a person can assume dominating control over another even if he or she makes no attempt to interfere directly in the choices made by the other. This kind of control, Pettit suggests, can be exercised through invigilation of a subject's choices. ³⁸ Where I invigilate your choices, I may be happy to let you choose as you wish and not interfere, but I am poised to step in and block your choices if they do not accord with my preferences as to how you should choose. Pettit suggests that where an individual becomes aware of such invigilation, it can lead to intimidation. This will in turn enhance the effect of the invigilation by giving the individual a reason to be cautious and deferential so as to avoid the likelihood that the invigilator will intervene to block or otherwise interfere with the individual's choices. Invigilation may involve the invigilator's physical presence. But it is quite possible – and perhaps a more effective way of ensuring compliance, as the person subject to invigilation can never be sure whether he is free of the other's attention – to invigilate choices remotely with the aid of various forms of technology.

Returning to the diary example, whether I become subject to this form of dominating power will depend on the motivations of the reader and the others who become aware of the information. If they are inclined to use the knowledge that they have acquired to their advantage, they may be able to invigilate my choices by, say, making me aware of the fact that they know what is in my diary and suggesting to me that they are unable to give me an assurance that it will go no further. In such circumstances, I may put in train various strategies to prevent further disclosure. I might ask what I can do to ensure that no one else finds out about the shameful revelations – in other words, submit myself to their will. Rather than tackle the issue head on, I might seek to ingratiate myself with them in the hope that they will not disseminate it more widely. I may simply keep out of their sight and hope that doing so will not lead to further dissemination and that any interest in me and the embarrassing/shameful information will eventually dissipate.

But even if those who have acquired this information have no interest in using it for their own ends, they may nevertheless possess dominating power. The knowledge that they have' exposes me to the possibility of interference – they have a power to interfere in my choices, which I do not control. ³⁹ Although they may have no immediate interest in using the information, the options that I have will vary according to any variation in their will. I will suffer no interference so long as they remain benevolent. But should that state of affairs change I might find that my options are changed for the worse. Republicans will say that the loss of privacy when my diary is read, and when the information disclosed to (or overheard by) others, means that they acquire dominating power, whatever their motivations or dispositions. This is so irrespective of whether or not I am aware that they have the information. As Celikates points out, awareness of one's dependence on the arbitrary will of others is not a necessary condition for domination. ⁴⁰ The simple fact that we are dependent – that others have acquired the power to interfere arbitrarily – diminishes our freedom. If I am unaware that I am being watched or that others have acquired information about me, the loss of privacy that I suffer is harmful to the extent that those who watch or collect the information acquire such power. Liberals will be inclined to say that the loss of privacy that we suffer in such circumstances is harmful – that people and institutions should not covertly observe or collect information about us. However, as we have seen, they will find it difficult to explain why the loss of privacy that we suffer as a result of covert surveillance or data collection is harmful – why privacy is valuable in such circumstances. While liberals are generally concerned about the effect that a loss of privacy will have on the autonomy of the subject, the focus of republican concern will be any unchecked inequality in power that is created by such a loss. Republicans will say the loss of privacy we suffer when others watch or acquire information about us is harmful to the extent that it provides others with power to interfere in our decisions that we do not control – the power to remove, replace or misrepresent options that would be available to us had we not suffered a loss of privacy. This harm arises whether or not we are aware that others are watching or acquiring information about us.

In 2013, it was revealed that intelligence agencies in Western democratic states had established various secret communications surveillance and data collection programmes, and had for several years been accessing and collecting the content of emails, conversations on social networking sites, telephone calls, internet browsing histories and capturing images from personal webcams (including sexually explicit images). The scale of these programmes is vast. It has been reported, for example, that intelligence agencies in the United Kingdom had tapped over 40 fibre optic cables, each of which has the capacity to transmit over a 24 hour period, an amount of data equivalent to sending the content of all books in the British Library 192 times over. The data that can be harvested from our communications and internet browsing histories are likely to present those who access it with information on: the number and nature of our relationships; the state of our finances; our political views, religion, sexual orientation, life plans and other aspirations; our opinions about others; our fears, predilections and foibles. It may, as the American Civil Liberties Union has pointed out, reveal that a subject uses telephone-sex hotlines, has contemplated suicide, is addicted to gambling or drugs, or has been a victim of sexual assault. ⁴¹ In short, they are likely to reveal more about us than is known by those who are closest to us. The more that is known about these things, the greater the likelihood that some opportunity could be identified to interfere in the subject's decision making in the various ways identified earlier – the information can be used to ensure that options that would otherwise be open to him are removed, or to manipulate his decision making. ⁴² If we know a subject's person's, strategies, fears and weaknesses and disapprove of these, we can devise counterstrategies, manipulate and nudge him in the direction of choices we think more desirable, or coerce him into taking options that we would prefer him to take. As Adam Moore points out, the acquisition of personal information exposes individuals to certain risks; that governments will use it to retain domination and expand its power, and that corporations will use it to ‘overwhelm individuals in a sea of solicitations and promotional advertisements or to control their employees’. ⁴³

Surveillance scholars and sociologists have recognized that surveillance provides the observer with power over the observed. ⁴⁴ Anthony Giddens, for example, has argued that in any form of social structure, whether it be the relationship between state and citizen, or employer and workforce, the collection and retention of personal information may be utilized as an instrument for ensuring the compliance of the sub-ordinate class – the citizen, or the worker. ⁴⁵ It gives rise, he says, to domination, and while he does not appear to think of domination as a necessarily ‘noxious’ phenomenon, it is obvious that when he speaks of domination in this context it is taken to be a harmful state of affairs. ⁴⁶ Privacy scholars have taken up this idea and suggested that we should think about the harm to which the loss of informational privacy gives rise in terms of the power that others acquire over us as a consequence. ⁴⁷ But those who have acknowledged the relationship between surveillance, loss of privacy and the acquisition of power appear to have recognized neither the importance of participation in political decision making as means of controlling power nor the role of privacy in facilitating such participation.

Securing freedom: Domination, privacy and political autonomy

In the discussion so far, it has been suggested that privacy can, in various ways, protect individuals from exposure to dominating power, and in doing so will provide the citizens of a republican democracy with conditions that will enable them to develop autonomous ways of life. ⁴⁸ In the final part of this section, the focus of discussion will shift to the issue set out in ‘Securing freedom’ section above – the means through which the citizens of a republican democracy can realize the ideal of self-government and secure conditions of freedom for themselves. It was suggested earlier that self-government requires ex ante participation in the political processes that generate norms that regulate the conduct of republican citizens – the nature and scope of laws, rights and entitlements. Privacy serves republican ends in a direct way by ensuring that others do have the kind of access to citizens through which they are able to acquire dominating power. But it also serves the interest in broad and effective political participation that all citizens hold in common. A republican right to privacy will have, to borrow a term used by Jurgen Habermas, a ‘dual character’. In serving individual interests it furthers a common good. ⁴⁹ A right to privacy serves individual interests by securing conditions of freedom that are required if citizens are to develop autonomous ways of life. Personal autonomy is, in turn, a pre-requisite for political autonomy – engagement in public life and, in particular, effective participation in norm-generating political decision-making processes.

Republican rights are not to be understood as pre-political norms – rights that establish the framework within which political decision making takes place. Rather, they are the product of political decision-making processes – processes in which citizens are expected to actively participate. ⁵⁰ There is in this sense, as Philip Pettit points out, no ‘temporal or causal gulf between civic institutions and the freedom of citizens’. ⁵¹ It is only through active participation in the practice of politically autonomous law-making, that citizens can develop a proper understanding of the legal order (the law, its actors and institutions), and determine both the content of the laws, and the nature and scope of the fundamental rights that they hold. ⁵²

The meaning and content of a republican right to privacy (in common with all other fundamental or constitutional republican rights) should, as far as possible, ⁵³ be a political settlement – an agreement that is the outcome of a well-functioning deliberative process. It is through participation in such processes that citizens can be said to exercise control over privacy-interfering conduct and themselves define the boundary between the public and private spheres. In other words, to secure for themselves an appropriate degree of individual privacy. An implication of this is that the account of the value of privacy provided here holds across political communities that may have differing views as to what aspects of individuals' lives ought to be regarded as belonging to the private sphere. ⁵⁴

Iseult Honohan suggests that in a republican democracy, rights ought to be considered fundamental if they are

necessary to sustain citizens as active participators in determining the social conditions of their lives, and to enjoy equally the common goods made through collective self-determination, as well as to give them access to the individual goods necessary to be self-determining citizens. ⁵⁵

For republicans, privacy ought to be regarded as a fundamental right not only because conditions of privacy may protect an individual from exposure to dominating power, but because such protection is also a pre-requisite for effective participation in deliberative processes that will determine the scope of a right to privacy. As Honohan points out, a life cannot be led entirely in public, and an agent's capacity to engage effectively in political life (or, perhaps, to engage in it at all) will depend on him having developed an autonomous way of life generally. ⁵⁶ The person who has fulfilling intimate relationships, a stable home life, and is satisfied with the progress that she has made in the projects that she has set herself, is likely to be far better placed to engage in political life, than the frustrated and dispirited individual who is labouring over the question of what kind of person they want to be and what kind of (private) life they want to lead, or who is unable to realize important personal (non-political) goals that he has identified as objects of a worthy and worthwhile life. Personal and political autonomy should be thought of as mutually supportive and sustaining conditions. Without the guarantee of privacy that protects an individual from dominating control, her ability to develop an autonomous way of life will be diminished, and in turn, her capacity for political autonomy will suffer.

If a republican democracy is to realize the goal of non-domination, then it will require a certain proportion of the polity to be autonomous political actors – to be suitably motivated and be capable of effective participation in political decision making. ⁵⁷ If what I have suggested above is accepted – that political autonomy is dependent on citizens having developed autonomous ways of life generally (personal autonomy) – and privacy is a pre-requisite for personal autonomy – then republicans will consider individual privacy to be fundamental to the republican project of securing freedom generally. To invert the idea – if all citizens are guaranteed the privacy that they require to develop an autonomous way of life, they will be better equipped (though not necessarily more motivated) to exercise political autonomy. The interests that all citizens of a republican democracy have in shaping a common future and realising conditions of freedom, and recognition of the role that privacy plays in this enterprise will lead to individual privacy is considered a pervasive and common good. Its value in promoting political autonomy cannot be considered independently of its value in providing conditions of non-domination that are a pre-requisite for personal autonomy, and vice versa.

In addition to promoting political autonomy indirectly in the manner I have just described, privacy also supports in a more direct way, the kind of political participation that is required to secure conditions of non-domination – one that is not grounded in the idea of personal autonomy. It was observed earlier that republicans will consider deliberative decision-making processes to be well functioning to the extent that they provide citizens with opportunities to present their points of view, and demand that proposals (and challenges to such proposals) be supported by public reasons. The effectiveness of such processes in establishing conditions of non-domination will depend upon a sufficiently broad range of views being canvassed, particularly the views of those, whose practices and beliefs do not attract majority support, and those from sections of the community that have suffered disproportionate and entrenched disadvantage. It will be necessary in order to secure the effective participation of those whose interests might be shared by a minority, lack popular sympathy or approval, or attract condemnation, to provide them with a guarantee of privacy.

Democracy requires that individuals be free, subject to certain narrowly defined exceptions, to express their political views and to present political arguments in public – to seek to persuade, and to criticise prevailing political practices and institutions. Effective political deliberation is constituted in various spheres of communication. Although active participation in political decision making requires individuals to engage at some point in public dialogue, deliberation among a more restricted group may be essential preliminary to this, and a right to privacy provides the boundaries that make such deliberation possible. ⁵⁸ As Annabelle Lever points out ‘individuals may be uncertain about the political worth of their views, the best way to present them, or their likely effects’. ⁵⁹ Privacy is valuable in this context because it affords individuals an opportunity to ‘develop and test their powers of persuasion and self-expression’. ⁶⁰ It provides those whose interests and views might be met with widespread disapproval, enmity, or outright hostility, space to reflect on the importance of those interests, seek the advice and reassurance of allies, to take informal soundings from those who might be expected to take up opposition, and consequently summon the confidence and fortitude to advance and have them subjected to critical public scrutiny. Here, the right to privacy serves the political process (and the aim of securing conditions of non-domination) by providing conditions that enable minority groups to determine how best to articulate their position and explore strategies for advancing it. Revelation of formative plans that are later abandoned or radically transformed may expose a political association to ridicule, or otherwise undermine its credibility, and the risk of exposure may have a debilitating effect on motivation to participate in public deliberation.

Conclusion

My aim in this article has been to identify the value of privacy in securing the republican ideal of freedom as non-domination, and in so doing, to articulate the justificatory grounds of a republican right to privacy. The issue of privacy has received little attention in the republican literature, but its value in protecting us from interference – for securing freedom as it is conceived by liberals – is obvious. In this respect, there is little that separates liberal and republican accounts of the value of privacy. Clear ground appears between the two where an agent's loss of privacy cannot be satisfactorily explained in terms of the effect of a loss of privacy on the autonomy of the subject. If the subject is aware of the loss, it is possible for liberals to explain the value of privacy by reference to the effect that losing it will have on the subject's capacity for authentic decision making. But this reasoning does not explain why we should consider privacy to be valuable where a subject is unaware that others have acquired information about him or are keeping him under observation. Although liberals can be expected to say that the subject suffers a loss of privacy as a consequence of such activity, they will have difficulty explaining why privacy is important in such circumstances, and why we might think that the subject of the loss has suffered some diminution of freedom. The republican concept of domination, however, provides us with the foundation of a broad and coherent account of the value of privacy that encompasses circumstances in which the subject (i) suffers interference as a result of the loss, (ii) is aware that he has suffered a loss of privacy, but suffers no subsequent interference and (iii) is unaware that he has suffered any loss of privacy, and suffers no subsequent interference.

The account of privacy provided here also reflects the republican belief that to be self-governing and to secure conditions of freedom for themselves, citizens must participate in the political processes that generate the norms to which they will subject. Liberals acknowledge that privacy is a pre-requisite for participation in democratic decision making – that it ‘helps to secure our legitimate interests in political choice and participation’. ⁶¹ However, an idea that does not find clear expression in liberal writing is that relationship between privacy, personal autonomy, political participation and the possibility of individual freedom. Effective participation in political decision making by individuals and groups requires a guarantee of privacy, and the only means through which citizens can be assured of privacy and other fundamental liberties is through political participation.