Privacy and the Importance of ‘Getting Away With It’

One reason people value privacy is that it allows them to do or think bad things – things that, if made public, would warrant blame, censure, or punishment. Privacy protects several types of freedom – and one of these is the freedom to be bad. This paper will argue that this is a good thing.

then we must have a legitimate interest in having (at least some of) our bad actions and thoughts remain private or unknown. I will argue here that there is a value in protecting our interest in being able to be bad and 'get away with it.' Giving an individual no choice but to be good or face condemnation can harm that person. Being able to hide bad actions and thoughts is important for autonomy, happiness, and the pursuit of non-moral goods.
Privacy's value, it is often said, lies in its power to protect people who have nothing to hide (Rachels, 1975:331). That privacy can also protect individuals who harm or wish to harm others is usually understood to be a serious reason for arguing against the protection or value of privacy. The arguments against protecting privacy have largely focused on security or protection against the wrongdoings of others. It has been argued, for example, that government surveillance of citizens is important for maintaining their safety and security (Himma, 2016). Feminist critiques of privacy further argue that, among other things, privacy as a concept has been used in the protection and sanction of domestic violence by promoting the idea that there should be no intervention in what men do in their private homes (MacKinnon, 1983). This might give the impression that, insofar as we care about privacy, we do so in spite of the fact that it can enable people to be bad. However, I will argue, in fact we often care about privacy because it allows us to be bad. People want to be able to have affairs; not give money to people in need; have inappropriate sexual thoughts; make offensive jokes with their friends; and have a reasonable expectation that they have some control over who, if anyone, finds out about these things.
The question I'm concerned with here is whether there is a morally defensible interest in the ability to keep bad actions or thoughts hidden. There is no doubt that some people have morally indefensible interests in keeping their bad actions hidden -anyone who has committed a crime has an interest in not being caught. However, I will talk here about having a 'legitimate interest' in privacy as meaning a morally legitimate interest.2 To 'get away with' an action is to succeed in keeping the action, or your part in it, hidden from a particular set of people -the police, the media, your partner, colleagues, friends -those who you believe would otherwise punish or blame you for having performed the action had this information not been hidden. If we understand privacy as the ability to control who has access to information about you, then 'getting away with it' is equivalent to being able to journal of moral philosophy 17 (2020) 416-439 maintain privacy over information that, if discovered, would result in censure of one form or another.
I will argue that there are cases where it is wrong to seek out and/or reveal the bad actions or thoughts of others -even where the only responses to the discovery of this information are appropriate and proportionate. This is not to be mistaken for saying that we should get away with all bad actions. Nor am I arguing that such an interest trumps all other interests -competing morally legitimate interests often need to be balanced against each other, and this is not a defense of the right to privacy or the right 'to do wrong' (see section 3). Violations of privacy may be justified where they are used to prevent or punish serious harms to others, for example. However, insofar as privacy is valuable, I aim to argue here that one of the things that makes it valuable is the fact that it does allow individuals to be morally worse than they would otherwise be without it. Our interest in having the chance to 'get away with it' should be factored into calculations about whether and when to protect informational privacy.
The paper will begin by setting out what I mean by having an interest in 'getting away with' being bad. Section 2 presents a series of cases in which we may feel intuitively that an individual's privacy should be protected because it allows that individual to be bad. In section 3 I will argue that intuitions about these cases reflect real reasons why being able to 'get away with it' is important -reasons including that it can protect autonomy and promote happiness. In section 4, I will consider one possible reply which argues that when reactions to bad actions are proportional and warranted, they are not strong enough to rob an agent of the freedom to act badly, so there isn't a strong case for protecting an agent's ability to 'get away with it. ' Two caveats are needed before I begin. First, some authors talk in terms of privacy rights and others in terms of an interest in having privacy protected.3 In arguing that we have an interest in 'getting away with it,' i.e. in being able to hide the fact that we have behaved badly, I will be presenting something that looks a lot like a right. Insofar as A has a legitimate interest in hiding wrongdoings, others do not have a claim against A to seek out or reveal information of her wrongdoing. The interest in maintaining the ability to hide wrongdoings is not one and the same as a right to privacy, but rather one reason among many in favor of privacy being protected, whether as a right or not.
A right to 'get away with it,' if a right, would be a component right to the right of privacy which shares a number of privacy's more complex features. For example, as has been highlighted with the right to privacy, the nature of the corresponding no-claim is complicated. Thomson (1975) gives the example of individuals who have a right to privacy relating to a marital argument, but points out that this right does not extend to ruling that someone who happens to be standing outside on the street overhearing the argument has violated the couple's privacy right in doing so. Similarly, privacy rights (if they exist) do not require that no one know our personal information, but rather that those who become privy to such information do not share it with others inappropriately (i.e. share it with those who are not entitled to it). If thought of in terms of rights, a right to hide bad actions also does not apply to all cases of information about these actions -if there is a reasonable expectation that this information will become 'public' (whether this is due to clearly marked speed-cameras, or the fact that the parties to an affair made no attempt to hide it) then other parties have not violated this right in sharing the relevant information.4 That said, I will avoid talk of rights here and stick with interests.
Second, it is also important to note that insofar as this is a partial defense of an individual's ability to be bad, the kind of bad actions and thoughts I am concerned with here are relatively minor ones. They do not include, for example, violence or serious infringements of the rights of others. As far as possible, I want to avoid a discussion of which actions are wrong, and in giving many examples, I hope that even if the reader does not think that they are all genuinely wrong, it is enough for my argument that the reader agrees that at least some of them are wrong. There is an important question about what types of bad actions should be protected and under what conditions, and how we balance this value of privacy against other competing interests. I address this briefly in the conclusion, but there is a great deal more to be said on this matter. The first step, however, is to convince the reader that there is a value in protecting the ability to 'get away with' being bad in the first place.

1
Why is Privacy Important?
Here are three reasons why privacy is valuable:5 (1) Privacy protects us from the wrongdoings of others.
4 See Tunick (2017) for a discussion of the conditions under which one can reasonably expect privacy. 5 There are many others -two relevant to the discussion below are autonomy and protection of good character which in some ways intersect these three reasons, though apply beyond them. (See, for example, Rössler (2005) and Allen (2016).) One argument for the value of privacy that does not easily fall under these headings is Fried's (1968) argument that privacy is necessary for intrinsically valuable ends and relationships: respect, love, friendship, and journal of moral philosophy 17 (2020) 416-439 (3) Privacy protects us from having our bad actions (and thoughts) discovered.
Many see the most fundamental value of informational privacy as being the protection it offers individuals from harm perpetrated by others, which is reason 1). Information is power, and we are all familiar with the ways in which access to private information about individuals such as religious worship or political party membership has enabled persecution by tyrannical governments.6 Access to private information also enables individuals to perform a variety of crimes: identity theft; violence from an abusive ex-partner (through knowledge of their ex-partner's current location); robbery (through knowledge of when someone is away from their home), etc. Individuals who have done nothing morally or legally wrong should nonetheless fear their information falling into the hands of those who desire to do them harm.7 Reason 2) can, but needn't, involve any wrongdoing.8 People often feel shame or distress if they are seen naked or engaged in a sexual act. This is true even if they were seen by accident, with no wrongdoing on the part of the person who viewed them. Protection from blameless shame would also cover an interest in not having medical information revealed.9 The shame or distress associated with reason 2) can explain why revealing such information is wrong.
Reasons 1) and 2) capture legitimate concerns -they are uncontroversial reasons why privacy should be protected. It is easy to explain why this is so, as the invasion of privacy leads to an individual's unwarranted suffering. Then there is reason 3). Consider the following case: Ellen wishes to do something that she knows to be bad. She believes that if others knew of this act, they would rightly blame and admonish her. She values her privacy because it is likely to protect her from this blame, allowing her to perform the wrong act. Ellen's privacy is important because of reason 3).
trust. However, even in this case it is the fact that privacy maintains the potential of 'getting away with' wrongdoings that in part constitutes its function in maintaining the possibility of trust. 6 See, for example, Moore (2016). 7 Note that the fear of others using your information for bad purposes needn't only relate to blameless information. Fear of blackmail based on information about your own bad actions is still concerned with the wrongdoing of others: Blackmail is immoral and illegal. 8 Several authors discuss protection from embarrassment, for example, as a value of privacy.
See Rachels (1975:323-4) and Nagel (1998:18). 9 Of course, sometimes reason 1) will also apply to these cases. One can be blackmailed on the basis of naked photographs, or unfairly discriminated against on the basis of medical information. However, this does not to rule out 2) being an additional reason in favor of protecting privacy.
Of course, when you have done something wrong, you rarely want other people to find out about it. No one likes punishment or condemnation. This, if it is a reason at all for protecting privacy, is hardly a good reason. Indeed, reason 3) is often understood as a reason against the case for protecting privacy. I agree that in many cases this is true. Nevertheless, I will argue, sometimes 3) is a legitimate reason to protect privacy.
In the following sections I will argue that there are bad actions, thoughts, and character traits that individuals have a legitimate interest in keeping hidden in order to protect themselves from appropriate levels of blame or condemnation. I will roughly organize these into three categories -deontological wrongs, utilitarian wrongs, and character wrongs (vices). In the case of Ellen, so long as the reader is left in a position where they can substitute something they believe is genuinely bad for "something that she knows to be bad" and have the rest of the vignette follow, then I will consider this to be a success for my case.
I am not the only person to argue that there is value in protecting privacy that enables individuals to 'get away with' being bad. However, these arguments differ significantly from the one I will present below. Scanlon (1975), for example, argues that we should not have the right to violate privacy on the (justified) suspicion of wrongdoing as such privacy violations give us access to information to which we are not entitled, even if we also get information about the wrongdoing to which we are entitled. In Scanlon's (1975:320) example, if someone steals an object of yours, this does not entitle you to train an x-ray device on their house to see where they are hiding it because in doing so you would be privy to information about them that they are entitled to keep private. Scanlon uses a device similar to the 'badness detector' to argue for the opposite conclusion to the one I will argue for here -he concludes that we do not have a legitimate interest in hiding wrongdoings, though we do have a legitimate interest in hiding other (non-blameworthy) information about ourselves which often trumps the rights of others to discover information about our wrongdoings: If it were the case that in looking into my house you would see only [the stolen] object and be blind to everything else, then the judgment that it would be wrong for you to take such a look into my house would be thrown into some doubt … It is important here that the second supposition -that in looking into my house you would see only your object, and nothing of my life, my possessions, my guests, etc. -is wildly contrary to fact. If things really were like that then your glancing into my house would not disturb any of the (legitimate) interests which motivate our conventions of privacy. Since allowing you such a glance would journal of moral philosophy 17 (2020) 416-439 advance another interest -yours, in finding your lost property -it would seem irrational not to allow this kind of investigative peeking. Scanlon, 1975:320 Unlike Scanlon, Tunick (2013) argues that one value of privacy is that it protects those who have done wrong. He argues that, when bad actions are known, punishment through public disapproval is often disproportionate to the violation in strength or length or both. "Informational privacy is particularly important for people who do have something to hide but who do not or no longer deserve punishment" (Tunick, 2013:650). At first glance this appears to fall under reason 3) for valuing privacy, but it actually fits better under reason 1). Punishment or condemnation that is disproportionate to an infringement, whether involving overwhelming social condemnation or inappropriate laws (for example where the death penalty applies for drug possession), demonstrates wrongdoing on the part of the punisher.
Here I will explore the prospect of a) an individual having engaged in activities that are blameworthy, where, b) if others found out about these activities, it would be appropriate for these others to blame or condemn the individual, and yet c) it would be wrong for someone to reveal information about the individual's wrongdoing -and d) it would be wrong for reasons other than those subsumed under 1) or 2).

What Should We Be Allowed to Get Away With?
An inability to keep bad actions hidden -which would be the case, for example, in the world of the 'badness detector' -would among other things, prevent people from being able to perform bad actions. This is not in the literal sense of being physically prevented, but rather in the sense expressed by Øverland (2007:308): "[T]o prevent something simply means to cause something not to happen or not to be done, or to be the reason why somebody does not or cannot do a particular thing." Waldron (1981:30) acknowledges that "[d]epending on the circumstances … merely telling someone that he should not do A may be a highly effective way of getting him to stop A-ing." The threat of punishment can prevent an action and as such 'prevention' needn't be understood as merely a physical barrier which happens before the aborted action. Social disapproval can be as motivating as the threat of punishment.10 Indeed, even the prospect of being blamed by just one individual whose opinion matters to you can be sufficient in preventing action. The power of social disapproval to prevent us from freely acting in the way we choose has been highlighted in other defenses of privacy, though the focus has been on actions that are "embarrassing" (Rachels, 1975:323), "unpopular or unconventional" (Fried, 1968:474-5). If fear of mere embarrassment is enough to prevent an individual from freely acting, there is no doubt that fear of warranted moral disapproval will have an equally strong effect.
In this section I will aim to show that the three main major ethical theories all generate cases that are both morally bad, and yet that, intuitively, you should be able to get away with. I will run through a number of examples of such cases. This is not intended to be an exhaustive list, nor do I expect everyone to agree on all the examples listed. My aim here is simply to stir intuitions in the reader. In the next section I will present the case for why it is important to protect the ability to 'get away with' such wrongdoings.

Permissible Harms: Utilitarian Wrongs
To borrow a case originally presented by Driver (1992): Roger and Bob are brothers. Bob has a failing kidney and needs a transplant immediately. Roger is the only available match -if Roger does not donate his kidney to his brother, Bob will die. Driver uses this example as an illustration of a suberogatory act, one which Roger is not required to perform, but he would be bad if he didn't perform.11 Departing from Driver's original example, imagine that Roger really does not want to donate his kidney. He is not close to his brother, does not think Bob is a nice person, and he is terrified of hospitals and operations. When Roger is tested to see if he is a match, his medical records remain privateit is up to him to reveal that his kidney could save Bob's life. Roger knows that if Bob and pretty much anyone else knew that he was a match they would (rightly) place a lot of pressure on him to donate and (once again rightly) deeply condemn him if he chose not to do so. Many of Roger's closest friends and family members would never speak to him again if they knew that he had willingly let Bob die. To avoid this Roger either "[N]onlegal punishment is hard treatment or unpleasantness meted out by ordinary people upon those they regard as blameworthy or in need of deterrence." Tunick (2013:648). 11 Driver (1992:287)  There are some acts or omissions which are wrong because they harm others and yet which, intuitively, we should be able to 'get away with,' because if we had no privacy concerning such acts, we would be robbed of the autonomy to make certain significant choices. Suberogatory acts -those not morally required of us, but which it would be wrong not to perform -such as Roger not donating his kidney to his brother, fall into this category. It is intuitively important to protect an individual's ability to choose not to perform some acts which would reduce the suffering of others.
Other such harmful acts that one should be able to 'get away with' are those which we have a corresponding 'right' to do. Waldron (1981) gives examples of such acts including spending all one's lottery winnings on champagne and racehorses, giving none of it to charity (thereby exercising property rights), and voting for a racist political party (exercising the right to vote freely).12 Many potentially immoral actions that have indirect bad consequences, cumulative bad consequences, or distal bad consequences, also fall into this category. For example, individual practices that do not mitigate harm to the environment (e.g. not recycling), eating meat, supporting unethical companies, buying clothing made in sweatshops, etc. We may do these things and believe that it is our right to do so, but also know that we should not be doing them and believe that if others knew of our actions it would be appropriate for them to blame us. As such these represent cases of bad actions where it is nonetheless valuable to have the ability to keep our performance of them private.
Utilitarianism is often criticized as being over-demanding. However, the fact that as a moral theory it requires a lot from us is not the same as concluding that we should be punished for or prevented from performing what utilitarianism rules to be bad actions. It could be true that it is morally wrong for an individual not to do everything in their power to increase welfare and decrease suffering, and yet it still be the case that it would be wrong to force them to always act in the morally best way possible. In fact, this is my point exactlythe very fact that an action is morally wrong is not always sufficient grounds for preventing individuals from performing it, and the fact that an omission is morally wrong is not always sufficient grounds for mandating it. Privacy acts to protect this middle ground.

Non-Harming Rule Violations: Deontological Wrongs
Sophie recently voted for a mayor who has promised to bring in strict regulations as to where you are allowed to place your rubbish. One day after these are brought in she forgets to put out her bins in time and they are not collected. Her bins are particularly full and smelly. Nearby streets have a different rubbish collection day, and some businesses on them have large open bins. One night Sophie sneaks over and places several large bags full of rubbish in these shop bins. What Sophie has done is bad -she has broken a rule which she herself endorses. She would not want other people to do what she did, and she thinks it would be right to blame her (even give her a fine) if what she did is discovered. However, Sophie and her family would have suffered if they had had to wait another fortnight for their rubbish to be collected, and her action is a oneoff. If she knew that she would be caught then Sophie wouldn't have done what she did. Because of this, there is a value in her having a good possibility of not being caught.
Insofar as one is moved by deontological considerations, there are numerous bad actions that come from treating oneself as an exception to moral rules which nonetheless we may feel are valuable to be able to 'get away with' doing.
In some cases there is little to no harm to others that follows from breaking moral rules, and considerable benefit to the agent in being able to break them. Let us adopt, for the sake of argument, a hedonistic or epicurean definition of 'harm' that takes a harmful action to be that which either reduces an individual's happiness or increases their suffering.13 On such an account, Sophie's action was deontologically wrong, yet it produced very little harm, if any. While the case against allowing people to get away with rule violations that harm others is simple, the intuitive appeal of restricting non-harming deontological wrongs is less clear.14 Some of these harmless deontological wrongs fall into the category of actions that are widely believed to be morally wrong, but which we do not believe 13 I will remain neutral on whether this is the correct definition of harm. However, it is interesting to note that in the privacy literature, the case of the 'voyeur' who violates privacy without the victim's permission or awareness is often represented as if they are harming the person they spy on in virtue of violating that person's right to privacy. 14 There is also a case to be made that individuals should be able to 'get away with' selfharming practices such as drug-taking or smoking, which are wrong on some deontological accounts.
journal of moral philosophy 17 (2020) 416-439 should be illegal -for example, petty lying or adultery.15 Others, like in Sophie's case, involve breaking laws. Many people feel this way about traffic violations (it might be wrong to park in front of a fire hydrant, but if an individual is not parked there for long, that individual had good reason to park there, and no one was hurt, he or she typically wants at least the chance to get away with it). We may also feel this way about sometimes calling in sick to work so we can have a day off, watching an amusing cat video at work, and abusing privileges (e.g. letting a friend use your pass to get into a museum for free).
Of course, if everyone performed such actions then there would indeed be a harm.16 In the case of minor crimes, for example, the occasional wrongdoing is only justifiable if the vast majority of such wrongdoings are detected and appropriately censured or punished.17 Indeed if there is no effort to hide bad actions, such actions which might only have been minor can become more serious as the very fact that they are not hidden can normalise or encourage such actions, making them more harmful, and by extension, morally worse. Where an individual commits a minor crime that is not harmful because it is an isolated incident, but which would be harmful if it became common practice, secrecy can be important in preventing the practice from being widely adopted.
The distinguishing feature of these cases is that the individual breaking the rule or law would not want it to be a general rule that these actions were allowed. It is agreed in general that people should be punished for traffic violations and not working when they should, and so in an all-or-nothing choice we would rather that no one could ever get away with such things than that everyone could do them all the time. But why should this be the choice? Having the possibility of being able to get away with breaking the rules sometimes, in cases where no one is harmed, can be genuinely valuable. 15 Adultery often does lead to harming others, but as on the definition of harm being employed here, this is only the case when the adultery is discovered, adultery remains harmless so long as it remains private. In other words it does not produce harm for as long as the adulterer continues to 'get away with it.' The "Ashley Madison" data breach in 2015 leaked the information of millions who used a dating website aimed at people looking to have an affair. If we feel that there was something wrong with such a data leak then this would be captured by reason 3). 16 It is important for the cases above that the individuals' actions are a one-off, representing special circumstances, as opposed to a mere contempt for those who stick to the rules. 17 Where there is a rule not to walk on the grass, for example, this is to protect the grass from damage caused by heavy footfall. One individual walking on the grass won't make any difference to its health, while hundreds of individuals will.

Vices and Bad Thoughts
Petra hates her colleagues and says nasty things about them to her family when she gets home from work. Her jokes often pick on features about which her colleagues are sensitive -their weight, age, personal circumstances -or they are mildly racist, sexist, or homophobic. If her colleagues ever found out that she was making such jokes, or overheard her, they would be (justifiably) very angry and upset. Petra knows that she is being bad when she makes such jokes, but she also takes pleasure in doing so (her family enjoy hearing them too), and it allows her to let off steam. If she knew that other people (particularly her colleagues) were going to find out about them, she would not feel comfortable making these jokes anymore.
Some actions are bad because they harm others, some are bad because they violate rules, but others such as Petra's jokes do neither.18 If one is a virtue ethicist or religious ethicist, vices represent moral failings. Yet, even if you believe that there is something morally wrong with maintaining or cultivating bad mental states, inclinations, and patterns of behavior, in many cases it seems that the life where one is able to pursue vices is a better one than that in which one is restricted to only living virtuously (in action, if not in spirit). Character flaws often invite judgment -sometimes moral, sometimes social. For this reason vices are often features that an individual wants to hide from others. Often vices manifest in behavior. For example, willingness to eat off dirty plates rather than bothering to wash them can be a product of laziness, while not revealing how much you earn when you insist on splitting the bill evenly with people you know earn a lot less than you reveals miserliness or meanness. That a vicious person wants to hide such behavior from others to avoid disapproval is perfectly consistent with that person believing that her behavior represents a character flaw which is worthy of disapproval. Nonetheless, she may also get pleasure from indulging in her vices and wish to have privacy so that she may continue to pursue them. Furthermore an inability to 18 For the sake of brevity I am assuming you cannot be harmed by something you do not know about, though see for example Benatar (1999) and Langton (2016) for arguments that derogatory jokes and hate speech made in private do harm others. If one adopts a non-hedonistic account of well-being, such as desire satisfaction or pluralist positions, then having your reputation damaged, your goals undermined, or being lied to, does reduce your welfare even if these things had no impact on your happiness or experiences. See, for example, Nozick (1974), Scanlon (1998), and Feldman (2004). journal of moral philosophy 17 (2020) 416-439 find times when she is unobserved so she can act naturally (viciously) may cause her anxiety and distress. The freedom to exercise certain vices in private extends further than just the ability to act in vicious ways privately. Technological advances are increasingly a threat to mental privacy. Current neuroimaging and behavior observation or tracking technologies can already reveal information about morally bad thoughts or traits, including implicit bias, lie detection, and inclinations to cheat (see Roskies (2016)). Certain thoughts might be wrong, or an indication (or indeed the manifestation) of the bad character of the thinker. This could be true even if they were not acted upon. The following thoughts or mental states may be blameworthy even if they remain unexpressed or unmanifested in action: someone wishes for the death of a hated colleague; has an explicit sexual fantasy about a child or a family member; or takes pleasure in hearing of a terrible act of war. In each of these cases the individual may know that they are a bad person for having such thoughts, and yet find it impossible to stop them. The discovery of these bad thoughts would make the individual's life worse because of something they had little to no control over. These too are cases where an individual may have a justifiable interest in 'getting away with' being bad.
It is often claimed that thoughts cannot be bad if they are not acted upon, but public reactions to 'bad' thoughts being revealed suggests that many people believe otherwise. In 2018 the actress Roseanne Barr posted a racist comment on Twitter, sparking considerable public outrage. Barr blamed her action on taking a sedative.19 Sanofi, the sedative's manufacturer, acknowledged that their drug had been linked to a reduction in inhibitions, and joked: "While all pharmaceutical treatments have side effects, racism is not a known side effect of any Sanofi medication." Sanofi and Barr could both be right -it could be both true that, had Barr not had racist opinions, she would not have expressed them while sedated, and yet she also would not have expressed such opinions had she not been sedated.
The question is what feature of this case was morally objectionable. That is to say, if we are right in condemning Barr in this case, is it because of her beliefs or her actions? If it is true that there is nothing morally wrong with merely having racist thoughts, then Barr's defense seems like a reasonable one (assuming for the sake of argument that she is telling the truth) -there is a way in which her mental privacy was undermined and she was not morally responsible for the action of posting the tweet. The case could be seen as analogous to that in which one says racist things in their sleep. However, Sanofi's reply reflected a view that was mirrored in other commentary, that having the racist beliefs was the problem. On this view the tweet merely revealed that Barr was a bad person, as opposed to Barr being bad because she posted the tweet. In other words, because her thoughts were bad, Barr would have been morally blameworthy whether or not she sent the tweet, though the public would not have known this, had she not posted it.20 However, it is still reasonable for someone who has bad thoughts that are not expressed to wish that such thoughts remain private.
Some vices bring pleasure and do little to no harm and some contribute non-moral values to an individual's life. Other vices, even when they remain secret, nonetheless make the vicious person's life worse, but they are not intentionally fostered, nor can they be overcome. For such vices the vicious person's life can be made 'doubly' worse by having these vices revealed. Protection of certain vices and bad thoughts through privacy can be valuable for these reasons.

Why 'Getting Away With It' is Important
There are several reasons why privacy is important insofar as it protects our ability to 'get away with it.' Here I will focus on just two of these. Autonomy requires having genuine choices, and having our choices limited to only good actions (or in the case of utilitarianism, very good actions) undermines meaningful autonomy. Being able to 'get away with it' also contributes to happiness both because being bad itself can bring pleasure or alleviate pain, and because the feeling that you could never hide instances of being bad can contribute to fear and anxiety. In many of the cases I have discussed above, the bad action or character trait, if it remains undiscovered, does not reduce overall happiness by an amount that is comparable to the reduction in happiness that would occur if it were never performed. 20 Some (Philips, 1984) do not consider racism to consist in attitudes that are not accompanied by action -rather acts are racist, and people are only racist to the extent that they perform such actions. On such an account, had Barr had negative attitudes towards other races that had never resulted in action that harmed members of such groups, Barr would not be racist. Benatar (1999) argues that this account fails to capture both the meaning that we attribute to the ascription 'racist' and the harm that can be done by being an object of contempt, even if is not expressed in actions, as it undermines something we care deeply about -our reputation.

3.1
Autonomy Informational privacy is important because it protects free choice and autonomy. Privacy protects autonomy both in giving an individual control over who has access to information about them (Westin, 1967;Rössler, 2005), and the ability to present oneself as they choose to different people (Rachels, 1975;Scanlon, 1975;Marmor, 2015). If I wish to avoid the judgment of others, but also wish to be bad in one way or another, a lack of privacy restricts my autonomy in two significant ways. First, I am now unable to act authentically -the fear of social judgment limits my choice of actions to those I would not naturally have chosen to perform. In this way, lack of privacy ensures that I am not able to live the life I choose in the way I want to, but must instead mold myself to social expectations. Second, privacy provides individuals with control over how others see them. I cannot present myself in the way I wish if I lack control over the information that others have about me.
The argument that privacy protects autonomy extends easily to the case of wanting to protect information about bad actions or thoughts. Without privacy protecting one's ability to 'get away with' being bad, one has their choice to be bad limited or removed, thereby undermining their autonomy. As I have argued above, just as the threat of punishment limits one's choices, so does the threat of social judgment or condemnation. It is not sufficient to argue that autonomy is not undermined in the world of the 'badness detector' because an individual technically could still be bad in such a world. Being physically able to act is not the same as being free to act. Consider the case of sodomy lawshaving laws that outlaw sodomy, limit one's freedom even though in the majority of cases individuals will not be physically prevented from engaging in sodomy.
In his 1981 paper "A Right to do Wrong," Jeremy Waldron argues that, in some instances, we have a 'right to do wrong.' I'm not arguing for such a right here, but there are clear parallels between the arguments of those who believe in such a right and arguments that could be given in favor of privacy as a protection for being able to 'get away with' being bad. Waldron's argument focuses on the value of autonomy which, he argues, would be undermined if we were only able to choose between morally good and morally 'indifferent' acts.21 According to Waldron (1981:36), if one is only free to choose actions when they are not morally required to do otherwise, then the only choices left will be "the banalities and trivia of human life," such as where to start shaving your face, or what 21 For replies see Galston (1983), George (1995), Øverland (2007), and Bolinger (2017). For further defenses of the 'right to do wrong,' see Waldron (1983), Enoch (2002), and Herstein (2012).
to have for dessert. Without being able to make meaningful choices, argues Waldron, we are denied personal integrity and self-constitution. There have been objections to this line of argument. According to Raz (1988:381), "[a]utonomy is valuable only if exercised in pursuit of the good." Øverland (2007) argues that even if allowing people to do wrong was necessary for protecting autonomy, that this autonomy is not more important than the burden placed on anyone who is harmed by the associated wrongdoing: "[W] rongdoers have no autonomy-based claim against being prevented when violating a general duty because autonomy does not count for more than the wrong" (Øverland, 2007:391).22 However, few of the objections to a 'right to do wrong' argue that we should have no opportunities to do wrong. Indeed, the fact that opportunities to do wrong remain, even when not protected by a right, has been used to support the argument that there is no case for establishing such a 'right' in any legal or formal context. Raz (1988:380-381) points out that we have plenty of opportunities to do wrong that follow from pragmatic constraints on possible measures to prevent them (he uses the rather dated example of being unable to stop someone from copying music from a cassette tape which is under copyright). George (1995:128) similarly argues that we have opportunities to develop autonomy by being bad, without needing a legal right to protect them: We can, I think, concede that certain important opportunities for integrity and self-constitution would be lost in a situation in which, despite the availability of a range of significant choices among morally permissible options, one had no opportunity ever to make an immoral choice. This concession does not, however, entail that there must be strong moral rights to perform immoral actions. Opportunities for immoral choice inhere in the human condition. They are, in a certain sense, ineradicable.23 Such arguments suggest that Raz and George, who both reject the idea of a 'right to do wrong' would nonetheless resist the world of the 'badness detector' 22 Øverland (2007) maintains that only actions that harm others are genuine wrongdoings. 23 Herstein (2012:359-360) rejects the suggestion that we have enough opportunities to choose to do wrong without protection in law. He argues that it is not sufficient for autonomy for one to be able to do wrongs by default (due, for example, to the difficulty or cost of enforcement), but that these must be rights protected by law for individuals to genuinely believe that they are (and, by extension, be) autonomous. I am not arguing here that there should be legal rights to 'do wrong,' but rather that, all else being equal, the interest in being able to 'get away with it' is one reason for protecting a legal right to privacy.
journal of moral philosophy 17 (2020)  where there are no barriers to the detection and publication of any wrongdoings. If they agree that such a world is undesirable because it goes too far in limiting autonomy, then it follows that they accept the point that protecting our ability to choose to do some wrongs is important. The belief that such autonomy should be protected needn't imply that one endorses the exercising of this autonomy. A further reason that privacy is valuable in enabling autonomy is that free choice is a foundation of democracy. In a democratic society one is free to vote for the party of their choice. This freedom is secured not only through the absence of coercion, but also because voting for political parties is secret. Reason 1) in favor of privacy, that it "protects us from the wrongdoings of others," is going to be a good reason in favor of secret ballots -perhaps you would be persecuted, attacked, or illegally discriminated against, if others discovered who you voted for, or forced, physically or through threats or blackmail, to vote a particular way. Secret ballots protect individuals from the wrongdoings of others. However, secrecy also serves another important function. Insofar as we would be put off voting for a morally bad party through fear of the bad opinion of others, secret ballots protect us not just from other people being bad, but from other people being reasonable -and in doing so they allow us to be bad.24 Informational privacy promotes autonomy by removing certain barriers to action. In the case of many bad actions, we feel that it is important for an individual to have the choice to perform such an action, even if it is blameworthy. Such freedom allows us to live more authentic lives, pursue non-moral goods at the expense of moral ones, and have control over the type of life we want to lead. It also serves to increase the freedom we have to exercise other rights which may require the choice to be bad to be fully realised, for example the right to vote entailing that an individual can vote for a racist political party and the right to private property entailing that an individual can choose not to give any of her vast wealth to help others (Waldron, 1981(Waldron, , 1983.

3.2
Welfare Engaging in certain bad actions and thoughts makes us happy. They contribute to a rich and enjoyable life. Morality can be demanding, and while moral values do and should play a significant part in our lives, they are certainly not the 24 Believing that a secret ballot is good because it protects an individual's right to vote for whomsoever they choose does require understanding voting as a right tied to individual autonomy. Mill (1861: Chapter x) rejects the idea of a secret ballot because he sees voting as a duty, not a right, and one that requires people to be accountable to others. For Mill, only under exceptional circumstances should voting remain secret. only values we have (Wolf, 1982). Privacy can be important in protecting our freedom to enjoy non-moral values over moral ones. Privacy insofar as it protects our ability to say and think bad things also enables individuals to "let off steam" (Nagel, 1998:20). In many of the cases discussed in section 2, there is little to be achieved by revealing another person's bad actions or thoughts. In revealing that someone has bad thoughts, for example, you do not make things better for people other than the thinker, and you certainly do not make things better for the thinker. In fact, it is likely that you make things worse for both. If I tell Petra's colleagues about the cruel jokes she makes, this will contribute to their suffering as much as hers.
Not being able to do certain wrongs can make us unhappy and decrease the quality of our lives.25 The feeling that, if you do wrong, it will be discovered, by itself contributes to anxiety, unhappiness and an overall reduction in welfare. According to Nagel (1998:5), exposure which results from a lack of privacy "would be emotionally traumatic or would inhibit the free operation of personal feeling, fantasy, imagination, and thought." For Nagel (1998:10), removing the ability to conceal 'bad' thoughts would cause "the inner life … to wither, or … require too much distortion." It is important to point out that almost everyone does have something to hide. Consider again the 'badness detector.' Knowing that any wrong performed would immediately be known (even if accompanied by the assurance that any reaction to this knowledge would be appropriate and proportional) would be sufficient to fear being exposed for past transgressions, and to feel that one should pause to scrutinize every single course of action before deciding what to do, to make sure one was acting in the best way possible. A 'badness detector' would contribute to a general feeling of anxiety. This could be true even for the person who genuinely does have nothing to hide -even where people try to avoid bad actions, they can recognize that they will be fallible in their assessment of which actions are morally permissible and which are prohibited. Consider the case of someone who knows that certain terms are offensive and therefore it would be wrong to use them, but who is also unsure of the latest consensus on which terms are offensive. This person, even if they only use the appropriate terminology, might live in fear that they will unwittingly use a slur to refer to a minority or subjugated group. This fear will be exacerbated when they cannot be sure that their use of these terms will not become public knowledge. 25 See Menninger (1968

Appropriate Reactions and the Limits of Privacy
I have argued here that an individual is sometimes wronged if she is not able to avoid the appropriate reactions of punishment or disapproval that result from others knowing of her bad actions or thoughts. In section 1 I was careful to distinguish between cases where privacy was valuable because it protected us from the wrongdoings of others, and where it was valuable because it protected us from having our bad actions (and thoughts) discovered. One potential objection to my argument is that the kind of punishment that would limit one's autonomy in the cases described above would not be appropriate, meaning that insofar as privacy protects an individual's ability to 'get away with it' this is valuable only because it protects that individual from disproportionate or unjust punishment -in other words, it protects the individual from the wrongdoing of others. Tunick (2013), for example, argues that information about an individual's bad actions and thoughts should be protected by privacy because in the age of widespread information dissemination via the internet, society often disproportionately punishes individuals for relatively minor moral infringements. Perhaps disapproval in many of the cases described above could be warranted, but it would be wrong to express it. Whether the 'punishments' in the cases discussed above are proportional or appropriate depends on the case we are considering and the type of punishment expected. In the case of rules with small but standard deterrents against breaking them (such as a fine for littering, or an official warning for wasting company time), it is hard to see how the punishment could be inappropriate except where the action should not in fact be prohibited. But this would indicate that the rule is a bad one, rather than that the person who violated it was bad in doing so.
More complicated is the case of social reactions. Disapproving reactions can vary considerably. If we say that an act of expressing personal moral disapproval is inappropriate, then we mean that there is something wrong in that expression (and the person on the receiving end of that expression is wronged by it). However, we should be reluctant to conclude that most of the social reactions that would prevent (or deter) individuals from acting in the above cases are inappropriate in this sense. This is because many of these 'punishments' are deeply connected to the rights and interests of punishers. Where an individual has cheated, for example, her husband would not be wrong to end the relationship upon discovering the affair. Where an individual discovers that someone has refused to donate his kidney to save his brother, or that someone has voted for a morally objectionable political party, she is not acting inappropriately if she ends her friendship with such individuals or refuses to invite them into her home. Where the action or thought in question is genuinely morally objectionable, it may be too much to ask that an individual show no outward signs of the blame or condemnation they feel, preventing them from performing the most natural expressions of moral disapproval (see Bennett, 2002:149).
The question of what kind of response to the discovery of someone being bad is appropriate is a complicated one, but in many of these cases it extends beyond mild and loving admonishment (though even where the response is nothing more than mild admonishment, this may be sufficient to deter people from acting). Think about some of the cases described in section 2 and consider if the following reactions are either too weak to be off-putting, or too strong to be reasonable: receiving a fine for using other people's bins, and having a general feeling of annoyance and distrust develop in your new neighbors; having your museum membership revoked because you lent your card to a friend so she could get in for free; friends not wanting to split a meal with you when they discover that, despite insisting on splitting the bill evenly, you have a lot more money than they do; having the object of romantic interest feel disgust towards you upon discovering your gluttony and lazy behavior; your siblings not allowing you to have unsupervised contact with your nieces and nephews upon discovering that you have pedophilic fantasies; a colleague asking to be moved to a different department when she discovers that you have been having sexual fantasies about her; your TV network discovering that you are racist, and taking you off-air; your family, on learning that you could save your dying brother by donating your kidney, applying serious pressure on you to do soyour parents will never be able to bring themselves to look you in the eye if you let your brother die; your friends discovering that, upon winning the lottery, you gave no money to charity, lose respect for you and return all the gifts that you got them with your lottery winnings; your children, upon finding out that you eat meat, determining to pester you until you become a vegetarian.
These are primarily cases where condemnation from your intimate social, romantic, or familial relationships will vary depending on the nature of that relationship, the characters of these people and the strength with which they hold the relevant moral principles, among other things. However, to deny that these reactions are morally permissible, meaning that they would fall under reason 1) rather than 3), is itself overly demanding. Furthermore, often we have an expectation of judgment even where it is not explicitly expressed. Women who drink while pregnant, for example, feel that they will be judged if they do so in public whether or not anyone says anything to them. As captured by journal of moral philosophy 17 (2020) 416-439 Banatar (1999:193) "we have interests not only in being treated with regard, but also in being well regarded."

Conclusion: What Does This Mean for Privacy?
Here I have argued that one value of privacy is that it protects individuals when they have a legitimate interest in being able to keep their bad actions and thoughts hidden. My aim has been to highlight a gap in the literature regarding the value of privacy, but this does not entail that privacy should be protected at the expense of the good reasons we may have for reducing privacy in some cases. Insofar as there is a value in being able to 'get away with it' that contributes to the reasons in favor of protecting privacy, this must still be weighed against any reasons to limit privacy -many of them relating to safety and wellbeing.26 I have argued that the interest we have in being able to keep our bad actions and thoughts private is morally legitimate, but this is only so in certain cases. For a start, it only applies to relatively minor transgressions. We do not have a morally legitimate interest in getting away with murder. One could argue that the very fact that privacy can be used to protect an interest in 'getting away with it' is not a reason in favor of protecting privacy, but rather a reason for reducing privacy. After all, one might wonder if the world in which a lack of privacy entails that individuals are not free to be bad would in fact be a better world. I am not sure if this is true. The extent to which privacy should be protected is a balance between competing factors. However, I do believe that even while there are cases in which an otherwise reasonable interest in 'getting away with it' would lead to more harm than good, a society where people altogether lacked the ability to 'get away with it' would not be a better society overall.
Take, for example, the case of privacy being used as a means of the oppression of women. It is important to acknowledge that a right to privacy has been systematically used in the past to protect particular kinds of wrongdoings -in 26 Another concern to note is that so far as having the ability to 'get away with it' is valuable on the grounds of autonomy, having access to information about others is often required in order to make autonomous decisions. Revealing an affair, for example, can give the aggrieved party the information to make autonomous decisions about whether he or she wants to continue their relationship. Thank you to anonymous 'Reviewer 1' for pointing this out. See also Tunick (2013:653). particular harms perpetrated against women in the domestic sphere (MacKinnon, 1983).27 This is most certainly not a defense of such practices. However, as much as the notion of privacy has been used to oppress women historically, it is also deeply important for protecting the interests of women.28 This extends to the protection of women's interests to 'get away' with being bad. For example, an ability to 'get away with' being bad can be particularly significant in areas relating to women's reproductive autonomy. Accessing services to procure a late-term abortion (see George, 1995:Ch 4;Herstein, 2012) is a possible case of wrongdoing which would require legal protection (a protection on the grounds of a broader non-informational definition of 'privacy'), but also a protection of informational privacy.29 The ability to keep information about one's decision to have an abortion private can be crucial in allowing many women to freely choose to abort. The value of being able to 'get away with' some bad actions outweighs the disvalue of these actions being prevented in all instances. In working out the balance between the importance of protecting informational privacy and the benefits of restricting privacy, we should look beyond our concerns about others being bad, and also factor in our interest in protecting our own ability to be bad and 'get away with it.'

Biographical Note
Cressida Gaukroger is a Departmental Lecturer in Philosophy at the University of Oxford and Tutorial Fellow in Philosophy at Jesus College, Oxford.

27
'Privacy' as discussed in this context is usually meant to mean something other than mere informational privacy. Men were protected from intervention which would have limited or prohibited domestic violence on the grounds that their right to privacy protected their ability to make what were regarded as fundamental personal decisions relating to the 'private' or domestic sphere. While this type of privacy is not the focus of this paper, one could raise concerns that protecting an individual's ability to 'get away with it' through keeping private information about such violations would end up protecting the same types of behavior. 28 Lever (2001Lever ( , 2005 argues that while the right to privacy has been historically misused to perpetuate gender inequality, this does not mean the right itself is at odds with equality. 29 One can object that this is not a case of wrongdoing. However, it is interesting to note, as pointed out by George (1995:112), that this right has often been defended by those who do believe that it is a moral wrong. Driver (1992:292) similarly argues that "obtaining an abortion for no compelling reason" could consistently be thought of as bad and yet permissible.