Reply to Critics

This article responds to the four contributors to the book symposium on Conscience and Conviction: The Case for Civil Disobedience. Those four contributors are Thomas Hill Jr, David Lefkowitz, William Smith, and Daniel Weinstock . Hill examines the concepts of conviction and conscience (Chapters 1 and 2); Smith discusses conviction and then analyses the right to civil disobedience and my humanistic arguments for it (Chapter 4); Weinstock explores democratic challenges for civil disobedience (Chapter 5); and Lefkowitz assesses the merits of a legal demands-of-conviction excuse for civil disobedience (Chapter 5). This ‘Reply to Critics’ addresses them in turn.

often to wave a white flag of surrender. In relation to some of their challenges, I resist that temptation, but, in relation to others, I have revised my views.
The four contributors did not coordinate their efforts as far as I know. Yet, felicitously, each considers a different part of the book. Hill examines the concepts of conviction and conscience (Chapters 1 and 2); Smith comments on conviction, but then analyses the right to civil disobedience and my humanistic arguments for it (Chapter 4); Weinstock discusses democratic challenges for civil disobedience (Chapter 5); and Lefkowitz assesses the legal excuse for civil disobedience which I call the demands-of-conviction defence (Chapter 5). I shall respond to each in turn.

Conviction
Hill explores, first, the conceptual terrain of conscientiousness as sincere moral conviction, and then the terrain of conscience as genuine moral understanding. I focus in this section on sincere moral conviction.
To give a bit of background, in my view, there are four conditions for a sincere moral conviction: (1) consistency, (2) universal moral judgment, (3) non-evasion, and (4) dialogic effort. Hill agrees that the consistency and universality conditions are plausible. After all, they have a respectable philosophical pedigree. That said, he notes that a familiar problem confronts the universality condition that we judge others and ourselves by the same moral standards. If the universality condition is interpreted strictly, then it rules out reasonable moral sensitivity to exceptions, and if it is interpreted loosely as a pro tanto constraint that does not specify the conditions for exceptions, then it gives only weak evidence that we are sincere when we claim that some act, such as going to war, would be wrong for us personally to do, but not necessarily wrong for other people to do. When questioned, we can satisfy the weak pro tanto universality condition simply by granting that there must be some (unspecified) relevant difference between us going to war and other people going to war. Moreover, Hill continues, 'Without some account of what [a person] regards to be relevant difference[s] between cases, we cannot even be confident that he is making a moral judgment.' 2 I acknowledge the force of this problem and can only say by way of reply that, when pressed, a genuinely conscientious person should be able to give the reason that, to her mind, distinguishes her case from others' cases; or rather she should be able to say why this act is particularly morally difficult for her to contemplate and why it might not have the same force for other people.
Concerning my third and fourth conditions for sincere moral conviction, namely, nonevasion and dialogic effort, Hill argues that these conditions are not necessary, but instead are 'evidently intended in part as conditions that the law might require as evidence for legal purposes.' 3 In essence, a court could look at both a dissenter's efforts to engage other people in a dialogue about her cause and her willingness to accept the consequences of her disobedience as evidence of her sincerity, which would distinguish her from other people who break the law.
To show that a person can have a sincere moral conviction without my third and fourth conditions, Hill gives the example of Thomas More as he is presented in Robert Bolt's play, A Man for All Seasons, where More aims simply to act quietly with consistency and universal moral judgment. In the play, More, who cannot support the Act of Supremacy that would make Henry VIII the Head of the Church of England, tries as hard as he can to remain quiet about it and to avoid the King's wrath. He does eventually display his convictions by standing trial and accepting first his imprisonment in the Tower of London and then his execution. But, Hill believes we can imagine a happier, parallel story for More, let us call him More-2, in which his opposition to Henry VIII is successfully concealed and he suffers no adverse consequences. It seems, Hill says, that More-2 has a sincere moral conviction that would not satisfy unqualified versions of my non-evasion and dialogue effort conditions. 4 There's a nice irony in Hill's use of Thomas More as a counterexample to my communicative account of conscientiousness since I open Chapter 1 of my book with a dialogue from A Man for All Seasons, and I claim that Bolt's version of More is a good example of someone who satisfies my four conditions for sincere moral conviction.
Hill's objection against the non-evasion and dialogic effort conditions for sincere moral conviction can seemingly be supplemented by two further cases. First, consider majority agreement. Suppose that Thomas More has a deep moral conviction that is shared by the majority of people in Henry VIII's England including the King himself, such as a conviction that subjects must have unswerving loyalty to the King. This More, call him More-3, will not be called on to defend his conviction in dialogue or to show his willingness not to evade the consequences of holding it since his view is the majority view. Yet, presumably, More-3 has a sincere moral conviction.
Second, consider social isolation. Suppose that Thomas More is not executed for treason by Henry VIII but instead is banished to a deserted island. Alone on his island, this More, whom we will call More-4, has no one with whom he can try to engage in dialogue about his convictions and no one to whom he can show his willingness to accept the consequences of his convictions. Yet, once again, presumably, More-4 has sincere moral convictions, such as convictions about how to treat the animals on the island, how to treat himself, and how to treat any person who happens to join him as well as more general convictions about how all people should treat each other.
Although the examples of More-3 and More-4 might imply that I concede to Hill that non-evasion and dialogic effort are unnecessary for sincere moral conviction, I do not concede that.
The response that Hill anticipates that I will make is to highlight the pro tanto nature of the non-evasion and dialogic effort conditions, which means that, in the happy-ending story of More-2, his conviction could count as sincere, Hill suggests, as long as More would have been willing to defend his conviction openly and accept the consequences if there had not been overriding reasons, such as the welfare of his family, that kept him silent.
I am not inclined to reply to Hill by appealing in this way to the pro tanto nature of the conditions. Instead, I would say that More-2 is not fully committed to his stance against Henry VIII's Act of Supremacy since he keeps quiet throughout; he is simply more committed to his family than to his belief about Supremacy. And, More-1 faces a conflict of commitments for much of Bolt's play between his love for the King and his commitment to his religion, and it is only when the conflict becomes intolerable that he demonstrates full conscientiousness by resigning his office and accepting his fate. In the book, I observe that it may be an implication of the consistency condition that, at any one time, we can have only one overmastering commitment that we would privilege before all others (32). This does not mean that we cannot have other deep commitments; we must simply hope they do not conflict in practice with the overmastering commitment that we privilege.
The reply that I am inclined to give to Hill has two parts. The first part appeals to a counterfactual test: would the person pursue dialogue and accept the consequences of his conviction if he had the opportunity to do so? In the majority agreement case of More-3, and the social isolation case of More-4, More's conviction can count as sincere as long as he would have been willing to defend it openly and accept the consequences if he were in a position to be challenged. It just so happens that neither More-3 nor More-4 is in a position to be challenged and so neither has an opportunity to show that he would meet the third and fourth conditions for sincere moral conviction. More-3 can communicate his view, if he wishes, but without much force since the majority agrees with him; and, More-4 cannot communicate his views to anyone.
That said, second, Hill's comments have prompted me to reflect further on the relations among the four conditions for sincere moral conviction. While I still hold that non-evasion and dialogic effort are necessary for it, I believe that they might be understood as corollaries of the first two conditions of consistency and universality respectively, rather than as independent conditions. They are important and non-obvious corollaries, and this justifies the effort to elucidate their features.
Consider, first, the non-evasion condition. Non-evasion derives from consistency. It is part of being consistent that we not adjust our conduct when doing so would be expedient but at the cost of the commitment that we purportedly espouse. Non-adjustment implies non-evasion. So, when Thomas More resigns his office, he shows that he will not adjust his conduct or evade the consequences even though it would be expedient to do so. His nonevasive response to his situation is a product of his consistency.
Consider, second, the dialogic effort condition. Dialogic effort derives from universal moral judgment. Part of judging other people and ourselves by common moral standards is being willing to try where possible to engage others in dialogue about conduct we believe is wrong. If we see someone doing something that we believe is wrong, such as a parent feeding Smarties to a baby, and we do not speak out, or if we are instructed to do something we think is wrong such as fire an employee without cause and we do not raise a question about it, then the universality of our moral judgment has no practical force behind it. It will not contribute to interpersonal and collective deliberations about how to act.
Moreover, the dialogic effort condition is meant to give us the kind of robust notion of a moral judgement that some act is wrong that Hill says we need. He says: The main point, all too briefly, is that a sincere moral judgment (as opposed, for example, to a judgment of law or etiquette) must not only be one that the person judging happens to be willing to apply consistently to others; it must also be a judgment that the agent takes to be defensible to others from an appropriately impartial point of view that includes the basic needs, interests and wills of others (See footnote 3).
A person's dialogic effort shows, amongst other things, that she believes her moral conviction is sufficiently credible that it can be given a reasoned defence (30). In other words, by seeking to engage others in a dialogue about the merits of her moral conviction, she shows that she takes her universally-applied moral judgement to be defensible to others who may not share her overall moral view.
Hill also considers whether my four conditions are actually sufficient for sincere moral conviction. He notes that my conditions are meant to leave room for mistaken moral convictions because those convictions deserve at least some respect and protection in a liberal society; but he says that my conditions need more content if we are to distinguish sincere but mistaken moral judgments from judgments about what is wrong in some nonmoral sense, such as contrary to rules of etiquette.
To make his point, he gives the example of a fastidious devotee to a certain kind of fashion or etiquette, who thinks it is wrong for her and for anyone else to wear scruffy blue jeans or to set a table 'incorrectly.' This person is sincere, consistent, non-evasive, and eager to engage in dialogue about her conviction, but hers does not seem to be a moral conviction, he says.
In my view, the lines between etiquette and morality are blurry at best since many, if not most, forms of etiquette are signals of inclusion, recognition, and respect for persons' humanity. In western societies, holding a door for someone, taking our shoes off when we enter a house where people remove their shoes, saying 'please' and 'thank-you,' and setting the table nicely for guests are all ways we show openness, receptivity, and respect. Other forms of etiquette might seem purely aesthetic, such as not licking our knives and not spitting back into glasses from which we've just drunk, but even those acts probably have a hygienic, safety-related respect-for-self lurking behind them. More generally, moral considerations of respect infuse the strict, socially-dependent, and culturally-varying norms that govern formal and ceremonial occasions including those that mark stages of life (christenings, bar mitzvahs, weddings, funerals), those that shape the processes of government (swearing-in ceremonies, inaugurations, openings of parliament, courtroom proceedings, state addresses), and those that honour the spiritual realm (religious services and observances).
But, these observations about etiquette do not touch on Hill's general point that we need some method to distinguish mistaken moral beliefs from beliefs about non-moral types of wrongness. This method will be difficult to supply since part of what matters for sincere moral conviction is that the person believes that the conduct in question is a moral matter. In short, when it comes to moral beliefs, we can be sincere but mistaken in at least two ways. First, we can be mistaken about whether the belief we hold is a belief that is genuinely relevant to morality, and, second, if it is relevant to morality, we can be mistaken about its value.
That said, in the book, I distinguish moral convictions from non-moral convictions such as factual convictions by arguing that the latter need not (and indeed should not) satisfy the universal moral judgment condition or, in turn, the dialogic effort condition. We can believe that the Earth is round and that people who believe otherwise are mistaken, without believing that they do moral wrong by being mistaken, and without our lack of moral judgment on them raising doubts about our own belief that the Earth is round. When we believe these things, then we have non-moral beliefs about the shape of the Earth. The same goes for non-moral social beliefs; we can believe that the appropriate way to acknowledge a speaker after a seminar talk is by clapping without believing that people who rap their knuckles on the table by way of acknowledgement do moral wrong and without our lack of judgement on them raising doubts about our own beliefs about the appropriate way to acknowledge a speaker.
I doubt that Hill's own proposed way to distinguish moral and non-moral convictions will be more successful. He says that, unlike a non-moral conviction, a sincere moral conviction must, in addition to having consistency and universality, be a judgment that the person takes to be defensible to others from an appropriately impartial point of view. But, of course, someone who holds a factual conviction that the Earth is round or that life on the planet evolved over millions of years holds a judgment that she takes to be defensible to others from an appropriately impartial point of view.
Before turning to Hill's reflections on conscience, let me reply to William Smith's comments on sincere moral conviction. He says first that, in my view, convictions are burdensome for the people who hold them: convictions are commitments that exert pressure on their advocates to communicate with others and to accept the personal consequences of doing so. Smith also says that convictions are burdensome on state officials, who must take steps to accommodate people's expressions of conscientious conviction. 5 Smith's summary of my view on this point is broadly correct, but requires some qualification. As we saw with More-3, a conviction is unlikely to be burdensome if it is the majority opinion. A conviction is also unlikely to be burdensome if it is a minority opinion that the majority thinks is innocuous, trivial, or otherwise non-threatening. Moreover, even if a minority-held conviction is deemed by the majority to be threatening, it will not be (unduly) burdensome to its defenders if they garner sufficient respect from the people who agree with them. 6 Like Hill, Smith objects to my dialogic effort condition. His complaint focuses less on whether my notion of sincere moral conviction is plausible and more on what this notion implies for my analysis of civil disobedience. Smith points out that the dialogic effort condition imposes motivational requirements beyond merely communicating a view since a dialogue is, in my view, 'a sustained purposive conversation or verbal exchange of thought carried out by two or more persons,' where there is a 'mutual orientation towards progress in common understanding' (217). This poses three problems for my analysis of civil disobedience, he says, all of which stem from the fact that civil disobedients do not embody such dialogic requirements merely in virtue of engaging in non-coercive, communicative dissent. In other words, for civil disobedients to have the distinctive motivations of parties to a dialogue, they would need to manifest a deliberative orientation toward a particular audience.
The first problem, Smith says, is tactics. In my view, paradigmatically, civil disobedients are suitably constrained in their methods; they show their dialogic ambitions by being neither unduly violent nor excessively coercive. Their self-restraint shows both that they realise they may be mistaken about their views and they desire to engage with people at the level of reason. Smith notes, however, that such self-restraint is not necessarily evidence of sincere conviction, modesty, and openness to dialogue. Instead, self-restraint can be used because it is instrumentally valuable and tends to secure a more sympathetic audience. Smith states that: Civil disobedience is not, to be sure, incompatible with dialogue and it is quite possible for citizens to aim at instigating a dialogue through this form of protest. The problem is that the aim of instigating dialogue cannot be inferred merely from the adoption of civilly disobedient tactics. 7 In reply, admittedly, we cannot know the contents of each other's minds. Therefore, if a dissenter lacks the kind of conscientious moral conviction that leads to self-restrained, nonevasive action, but nonetheless behaves outwardly in identical ways to a conscientious dissenter, we have no choice but to treat the two alike, as there are limits to how much we can or should seek to inquire into each other's minds. 8 That said, most likely, when a person lacks sincere moral conviction, her behaviour will reflect that, and will differ from the behaviour of someone with sincere moral convictions. By looking at behaviour, we can test, roughly, whether a person displays the consistency, non-evasion, and dialogic effort features of sincere moral conviction. We can ask: is she standing up to be counted for her view? Does she try to engage others in dialogue about her view? How much does she benefit personally from championing this view? How close is the connection between any personal benefit and the actions she is willing to take to defend the view? For instance, is championing environmental protection serving her well to win election? For how long has she shown a commitment to environmental protection issues?
The second problem, Smith says, is that associating civil disobedience with the dialogic effort condition may lead public officials to have unreasonable expectations of civil disobedients, since dialogue requires not just self-restraint, but a complex set of attitudes associated with deliberative interaction (see footnote 7). If dissenters are unwilling to engage with officials, but willing to engage with other groups, then authorities might decide that they are not conscientious and, hence, deserve no special protection.
Smith's point is well taken, but it is also purely pragmatic. It simply shows that disobedients must be as clear as possible about why they are unwilling to engage with officials when they are unwilling to do so, and they must reflect carefully on whether their unwillingness is conscientious or self-protective.
The third problem, Smith says, is that the dialogic effort condition is unnecessary to show that civil disobedience is more conscientious than personal disobedience. All we need in order to show this is a weaker communicative requirement short of dialogue whereby civil disobedients simply aim to communicate their convictions to an audience by appealing to the audience's sense of reason or justice. They can do that without the demanding attitudes of reciprocal concern and equal recognition that distinguish a dialogue.
Smith anticipates my reply, which is to note that dialogic ambitions signal that people appreciate that they might be mistaken and, hence, they are open to hearing what others have to say in response. Additionally, as noted above, they signal that they believe their views are sufficiently credible that they can be given a reasoned defence to others who are given the space to challenge them in reply.
To this, I would add that showing that civil disobedience is more conscientious than personal disobedience is a by-product of the communicative account of conscientiousness. The fact that communication short of dialogic effort might suffice to show that civil 7 Smith (2014), § on The Demands of Dialogue. 8 In his PhD research at ANU, Ten-Herng Lai argues that civil disobedience need not be conscientious in the sense I describe here. disobedience is more conscientious than personal disobedience (on a weaker notion of conscientiousness) is irrelevant.
Smith argues that, contrary to my claims, sometimes being unwilling to engage in dialogue can be evidence that we believe our convictions are credible: our unwillingness 'conveys to others that we treat our convictions as so firm and settled that there is no benefit to subjecting them to further critical scrutiny (see footnote 7). Christopher Cowley makes a similar point: Rosa Parks was also not open to the possibility of error in her conscientious conviction. Nor was Thomas More. Both were open to dialogue in the minimal sense of being willing to explain their position to anyone who sincerely wanted to know; but they were not open to dialogue in the sense of accepting the possibility of being mistaken. It would make sense for Rosa Parks to ''try to understand'' her racist interlocutors from the point of view of strategic planning, but it is essential to the nature of her unperplexed moral conviction that she would not even try to understand her interlocutors from a moral point of view. 9 In reply, Smith and Cowley are correct that Rosa Parks's (presumed) unwillingness to debate the merits of her conviction could be taken as evidence that she believed her conviction was credible. But, we should not overstate the nature of such unwillingness. If conscientious, Rosa Parks would have tried to understand her opponents' motivations, underlying commitments, upbringing, and education in that spirit of Gandhian non-violent, non-hating resistance that distinguished the movement that Martin Luther King Jr led. She would also have sought to see things from her opponents' perspectives with their attendant fears, misperceptions, and cultural pressures. She would have sought to know what common ground there was, if any, between their outlook and hers in order to move toward mutual understanding. As John Stuart Mill writes in On Liberty: [T]he source of everything respectable in man either as an intellectual or as a moral being [is], namely, that his errors are corrigible. He is capable of rectifying his mistakes, by discussion and experience. Not by experience alone. There must be discussion, to show how experience is to be interpreted. Wrong opinions and practices gradually yield to fact and argument: but facts and arguments, to produce any effect on the mind, must be brought before it. Very few facts are able to tell their own story, without comments to bring out their meaning. The whole strength and value, then, of human judgment, depending on the one property, that it can be set right when it is wrong, reliance can be placed on it only when the means of setting it right are kept constantly at hand. In the case of any person whose judgment is really deserving of confidence, how has it become so? Because he has kept his mind open to criticism of his opinions and conduct. Because it has been his practice to listen to all that could be said against him; to profit by as much of it as was just, and expound to himself, and upon occasion to others, the fallacy of what was fallacious. Because 9 Cowley, Christopher, 'Conscientious Objection and the Limits of Dialogue.' Submission to a special journal issue edited by Maeve Cooke and Danielle Petherbridge (under review). C.A.J. Coady also raises this issue briefly in his review: 'Someone's categorical, unqualified rejection of slavery or torture can appear to lack the questing, openness to revision, that Brownlee's ''conscience'' apparently requires. Yet, those few intellectuals who stood out against slavery in the past and those other few who rejected the Western apologists for torture in the war on terror, were clearly conscience-driven, even if rigid in their (I believe correct) convictions. It is precisely this unfazed strength of conviction that often seems to indicate the presence of conscience. he has felt, that the only way in which a human being can make some approach to knowing the whole of a subject, is by hearing what can be said about it by persons of every variety of opinion, and studying all modes in which it can be looked at by every character of mind. No wise man ever acquired his wisdom in any mode but this; nor is it in the nature of human intellect to become wise in any other manner. 10 The committed dissenter must walk a tightrope between having sufficient confidence in her view that she is willing to bear the risks of defending it through suitably constrained, communicative disobedience, and having sufficient modesty about her epistemic position and moral understanding that she is genuinely open to hearing competing views in a spirit of respectful, dialogic engagement. If her views gain traction over time, then she may feel more secure in her adherence to them, but must still appreciate Mill's important dictum that we not presume that we are infallible. 11

Conscience
Returning to Hill, after exploring the notion of sincere moral conviction, he turns to the notion of conscience. He offers a rich overview of four accounts of conscience in the literature, and along the way makes a few critical comments about my conception of conscience, which I situate within a pluralistic moral framework.
First, he notes that, whereas I emphasise honing our practical moral skills, including our emotions, he would emphasise 'developing and exercising our rational capacity to understand basic moral requirements and to apply these intelligently to particular cases.' 12 He also thinks that I am mistaken to hold that, through conscience, we can know what our special responsibilities are in a morally pluralistic world.
Here, Hill and I must agree to disagree about which activities properly cultivate conscience, and consider whether studies outside philosophy in psychology and neuroscience provide evidence to support our claims. Also, I would note that, although conscience may only ever give us an imperfect understanding of our special moral relationships, it will give us a much better understanding than we would have without it.
Second, Hill argues that monistic moral frameworks can accommodate more of the ideas that I credit to a pluralistic moral framework than I acknowledge. Briefly, he says that 'moral theories that seem to offer a single first principle are often more complex than they initially appear' and that 'although Butler and Kant do not affirm all of the features of Brownlee's moral pluralism, a monistic moral theory is compatible in principle with many of them' (see footnote 12). That said, he then closes his remarks by acknowledging that: … although monism of at least a modest kind (as suggested above) remains an aspiration for many of us who work on normative ethical theory, I think that there was wisdom in Rawls' remark that moral pluralism is the default position (see footnote 12).
Unsurprisingly, I agree. Moral pluralism makes appropriate space for tragedy, moral conflicts, remorse, and regret while at the same time leaving room for us to cultivate genuine moral skill and understanding. As Candice Delmas notes, we must, first, do our best to see things as they actually are, which includes seeking out information. Second, we must try to resist self-deception. Third, we must try to understand other people and learn about their experiences. 13 In these ways, we can cultivate the genuine moral understanding that constitutes conscience.
Let's turn now away from conceptual matters to the moral and legal qualities of civil disobedience.

Humanism and Rights
In my view, people have a moral right to engage in civil disobedience regardless of either the cause they champion or the regime in which they champion it. This right is rooted in a humanistic respect for people as reasoning, feeling, and expressive beings who are capable of deep commitments. In short, when a person has a deep moral conviction, she has strong interests in having some freedom to act expressively and conscientiously to manifest that conviction in her life and to deliberate with others about its merits.
Smith doubts whether humanistic respect can ground a general, content-insensitive moral right to civil disobedience. He says that, if the egalitarian, respect-driven demands of humanism apply to all parts of our behaviour which negatively affect others' rights, agency or dignity, then humanism cannot support a right to civil disobedience that advances antihumanistic goals such as bigotry, sexism, racism, or xenophobia. He asks: if a neo-Nazi engaged in suitably constrained civil disobedience to champion his cause, would a judge really properly owe him an apology if she were to breach his moral rights and punish him?
In reply, admittedly, such an apology could seem strange and confusing. But, the judge would simply have to be clear about what she was saying, namely, that the apology was not intended in any way to excuse or condone the neo-Nazi's cause. Rather, the apology was given strictly for not being able in this case, for whatever reason, to make some space for the person's suitably constrained, civilly disobedient expression of his conviction.
Moreover, Smith's point highlights an inescapable tension in liberalism, which lies in being tolerant of the intolerant. In my view, the humanistic principle protects the civil disobedient who has a bigoted cause, but not the state that adopts a bigoted policy, for three reasons. First, most likely, the state's bigoted policy would not be suitably constrained, but instead would have large-scale ramifications that signal or foment a broader prejudicial attitude about certain peoples. 14 Second, states are not people. They do not have the expressive interests that reasoning, feeling beings have to be given some space to express their views. Third, the closer we get to the centres of political power, the more restrained our methods must be.
Smith's objections lead him to raise a final point about the kinds of protection we get from our moral rights. In my view, moral rights of conduct give us defeasible protection against any form of interference by any party. Smith rejects this on the grounds that, as an unlawful action, civil disobedience naturally triggers interfering responses from law enforcers, such as police arresting protesters for their unlawful actions. Thus, he says, (reasonably good) public authorities may have greater moral leeway to interfere with civil disobedience than I acknowledge. 15 To support his point, Smith notes, first, that some rationales for interfering with a rightsprotected act do not threaten autonomy in the way that deterrence-driven interference does. For instance, a financial penalty might be imposed on disobedients to get them to contribute to the societal costs of their protests. Second, he observes that, since protesters often intend for their conduct to culminate in arrest, their rights are not overridden by police when police arrest them, and police owe them no apology for doing so. Third, Smith says that it is difficult to reconcile my model with the type of interference that is 'permissible' on my account. He says that it is striking that, in my view, the right to civil disobedience can be overridden by grounds for punishment.
In reply to Smith's first point, I agree that modest financial penalties aimed at covering costs do not necessarily impinge on disobedients' personal freedom. When imposed for that reason, and known to be imposed for that reason, modest penalties do not actually interfere with the right to civil disobedience. In reply to Smith's second point, I would note that a person might intend, through the exercise of free speech, to be charged with hatecrimes or obscenity in order to test the law, but that does not mean that her freedom of speech hasn't been interfered with when she is so charged. Similarly, protesters might intend to be sent to jail, but that does not alter the fact that being sent to jail intrudes upon dissenters' moral right against punishment. Finally, in reply to Smith's third point, punishment of civil disobedience is not morally permissible. An apology is owed to civil disobedients, which means that punishment of civil disobedients is, at best, wrong but justifiable.

Democracy
Another objection made against a moral right to civil disobedience is that civil disobedience is undemocratic. This objection, which I call the democracy problem in the book, has three aspects: (1) The epistemic problem that, ostensibly, civil disobedience challenges the democratic legislature's supreme right, and epistemic qualifications, to take strategic decisions for the whole community (174). (2) The paternalist problem that civil disobedience is disrespectful because, in engaging in it, we assume that we know better than other members of our community and we give ourselves license to breach laws that others are following (174-175).
(3) The procedural fairness problem that civil disobedience unfairly disregards the procedural democratic mechanisms that the society has adopted to resolve disagreements (175).
In the book, I respond, first, that legislators are not invariably, or even usually, better placed or better motivated than on-the-ground experts to account for all of the salient reasons for and against a policy (175-176). Moreover, even if they were, they could still benefit from concerted minority opposition, as it keeps them alive to all of those salient reasons (176). Second, democracy not only accommodates diverse positions, but also is sensitive to shifts in perspectives. It is a work in progress through which defeated parties and persistent minorities can continue to challenge majority views through means such as civil disobedience and sometimes, thereby, rectify democratic deficits by drawing attention to neglected issues (176). 16 Third, as David Lefkowitz argues, persistent and vulnerable minorities endure the comparative unfairness of having a lesser capacity to get their concerns heard and to influence policy choices before formal decisions are taken. Given this unfairness, there must be some leeway in how such persons are allowed to participate (177). According to Daniel Weinstock, my efforts to show that civil disobedience is not undemocratic fail for two reasons. First, they misconstrue the epistemic claims that legislators and disobedients may make; and second, they conflate a concern for persistent minorities with a concern for outvoted minorities. The unifying explanation for why my rejoinders to the democracy problem fail, he says, is that I adopt 'an excessively thin set of presuppositions about democratic practice and about democratic institutions.' 17 He observes that: … well-functioning liberal democracies are valuable and fragile accomplishments, and those who would presume to disobey their laws simply because they disagree with them are guilty of a form of moral self-indulgence, no matter how deep their convictions are that the laws that they are disobeying are wrong. 18 In reply, it is an empirical question whether well-functioning liberal democracies are fragile accomplishments that would be threatened rather than strengthened by people resorting to morally misguided, but suitably constrained, communicative breaches of law to defend their convictions. Since democracies, once established, tend to endure and since many eccentric people living in democracies protest at the edges of the law, liberal states are probably less fragile than Weinstock suggests. Indeed, when they give people some space to engage in civil disobedience, they not only show humanistic respect for those people, but also produce the side effect of releasing tensions that might otherwise explode in more radical forms of protest.
Furthermore, Weinstock fires at a wider target than he intends. His charge if it were to hit home would strike not only civil disobedience, but also private disobedience, such as the religious doctor who quietly refuses to perform abortions for rape victims and who, I stipulate, does not disobey in response to a gross injustice. Although one might say that the doctor is indeed being morally self-indulgent, that is neither the common view nor Weinstock's view, I suspect, of principled, but misguided private disobedience. (In contrast with the common view, I do say that such a doctor is probably morally self-indulgent since he is evasive and unwilling to engage others in deliberation about his view.) Finally, I disagree with Weinstock's suggestion that civil disobedients who are not responding to gross injustices are morally self-indulgent. Paradigmatically, civil disobedients are sincere, serious, conscientious actors, who appreciate that they might be mistaken in their views and who wish to engage with others in reasoned debate. They do not disobey the law because they disagree with it. They disobey the law because they believe the law they oppose is morally wrong and should be lastingly changed. 19 Weinstock anticipates some of these responses. He also analyses the specific responses I give in the book to the three parts of the democracy problem listed above. Let's take each in turn.
Concerning the epistemic problem, Weinstock argues that I do a disservice to legislatures since, despite political parties' ideological commitments, serious political parties have good grounds to assert their epistemic superiority over individuals, as they develop elaborate policy platforms that are the products of fairly complex and lengthy studies, party debate, and consultation. Parties further develop their epistemic credentials through their work in government commissions, inquiries, electoral debates, and town-hall meetings.
In reply, despite their credible efforts to attain epistemic superiority, serious political parties still face epistemic threats as well as motivational threats that stem not only from their ideological commitments, but also from the pressures of powerful interest-groups, lobbyists, and their own self-preservation, all of which can make it difficult for them to acknowledge or use the knowledge they may well have.
Moreover, as C.A.J. Coady argues in his review of my book: … [E]ven if we attach particular importance to institutional and communal authority, an individual's membership of many institutions and communities, other than the government and nation, means that a given individual confronts epistemic authority from several sources and needs to resolve possibly divergent information from each of them. This is further support for Brownlee's claim that certain experts can be in a better epistemic position than government decision makers; she mentions environmentalists and soldiers regarding environmental protection and warfare respectively, but might have mentioned skilled investigative journalists, international lawyers, exdiplomats, and many more, as well as those who rightly respect some of these authorities. 20 Coady then notes that epistemic standing is not only about access to, and understanding of, factual information. It is also about moral information. He states: If you think that there is such a thing as moral authority or expertise, then it is no more likely to be found in a democratic legislature than in many other places in the community (or various sub-communities); if you believe in moral authority or expertise but not in its communal variety then again there is no principled superiority of state authority to private judgement; if you do not believe in moral authority or expertise at all, then there is even less reason to defer to the moral backing that the rulers have for their decision without close personal attention to its basis. 21 I wholeheartedly endorse these points. Weinstock then argues, following Jeremy Waldron, that epistemic superiority is not the only, or even most important, reason for citizens to defer to legislators. The most important reason is coordination. Democracy is about finding policy solutions in the 'circumstances of politics' where there is every reason to believe we will continue to disagree, but nonetheless must find some policy that we can all live with. 22 Democratic legislatures are best placed to play that coordinating role. By allowing everyone's view to be part of the democratic calculus-through the vote-they show respect without requiring anyone to change her views. By contrast, Weinstock says, those civil disobedients who are not responding to gross injustice give their own views preeminent status over others' views, seeking to press others to change their views.
The comparison is inapt since civil disobedients are not in the business of coordinating, nor should they be. Legislatures and disobedients play very different roles in the democratic formation of policy and law. In well-functioning liberal democracies, disobedients are not law-makers but gadflies. 23 Weinstock anticipates this response and accepts it, but insists that civil disobedients only perform a legitimate gadfly service (as a last resort) when they raise relevant issues that are not being considered at all by legislators. It's not enough, he says, that disobedients raise an issue that they believe should receive greater attention.
I acknowledge that there is no guarantee that civil disobedients will defend the causes that most need to be given greater democratic attention: we have to hope that citizens in a democracy are sufficiently attuned to the most important issues that some of them will protest when an important issue is overlooked. But, civil disobedients not only raise an issue that they believe warrants greater attention. They also model for others a suitably constrained mode of dissent, which is part of their gadfly service. They opt to be a gadfly, not a scorpion.
Turning next to the paternalism problem, Weinstock notes that civil disobedience is not the only way to keep democratic deliberation alive between elections. Lawful demonstrations, lawful strikes, letter-writing, and blogs are all ways to engage in political argument and contestation. Weinstock says that I acknowledge this and that, consequently, I make a concession that is damaging to my overall position. I say that, 'when breakdown in the mechanisms of engagement occur, civil disobedience is one comparatively modest way to rectify these democratic deficits by focusing attention on neglected issues' (176). Weinstock takes this claim to imply that when the mechanisms for engagement have not broken down, most civil disobedience is illegitimate and, therefore, only remedial, injustice-fighting civil disobedience is legitimate. But, my claim does not imply that because my case for the existence of a content-insensitive right to civil disobedience does not rest on its democratic credentials. Instead, my case for the right to civil disobedience rests on humanistic respect for persons as reasoning, feeling, expressive beings (discussed above). The democracy-compatibility of civil disobedience simply shows that we need not worry about the moral right to civil disobedience on democratic grounds.
Next, Weinstock looks at the procedural fairness problem and my claim that what it means to be a comparatively disempowered minority is context-sensitive: people who are comparatively empowered relative to most other people can be comparatively disempowered in contexts where their power is inadequate. For instance, the members of minority parties are not comparatively disempowered in relation to most members of society, but they are often comparatively disempowered in relation to the governing party.
Hence, a breach of formal norms, such as a walkout, may be the only effective way for them to influence policy formation (177).
Weinstock argues, first, that minority parties are given more space to contribute than I acknowledge, on government committees and in more informal ways, and, second, that they have no right to have their views prevail since they are, after all, in the minority. 24 My claim is not, however, that they should prevail, but that they should have an adequately large microphone to voice their views. Weinstock says that liberal democracies have mechanisms to ensure that minorities can voice their views. But, of course, there is often a difference between rules and practice, and powerful bodies can use the rules to their advantage unless they are checked. Sometimes, the only way to check powerful bodies is through non-conventional or (suitably constrained) illegal means.

The Demands-of-Conviction Defence
The moral right to civil disobedience underpins a legal excuse for civil disobedience, which I call the demands-of-conviction defence. Lefkowitz agrees with me that civil disobedients should have access to such a defence, but he believes that it is best understood as a justificatory defence, not an excusatory defence.
Lefkowitz begins by exploring the conceptual terrain of justifications and excuses, which I traverse in Chapter 5. He argues that when we give a legal excuse for a breach of law we concede that we 'ought to have recognized the criminal law's authority but argue that [we] should not be blamed or faulted for [our] failure to do so. ' We admit an error in reasoning but contend that the error was reasonable, meaning that we lived up to the standards that could reasonably be expected of us in the circumstances. 25 By contrast, Lefkowitz says, when we give a justificatory legal defence, we are asserting that the state lacks authority over us in the circumstances, and hence we did not act wrongly by failing to follow the law. 26 According to Lefkowitz, in saying this, we are not 24 In 1987, Frank McKenna's Liberal Party won a landslide victory in New Brunswick, Canada, securing every one of the 58 seats in the provincial legislature. Admirably, the party put in place various measures to respond to the internal proceedings of the Legislative Assembly and to ensure meaningful debate. They made arrangements for Liberal backbenchers to ask questions of the Government. They seated some cabinet ministers, including the Premier, to the left of the Speaker. They provided research assistance to the unelected opposition parties. Eventually, they allowed the leaders of the unelected opposition parties to sit on the floor of the Chamber in seats normally reserved for the media. In all of these ways, the Liberal Party gave minority parties exactly the kind of space that Weinstock says minority parties have in liberal democracies. The point to note, though, is that McKenna's Liberal Party was not legally required to do these things. They acted remarkably well by implementing these practices, but another party could easily have taken the view that they had a mandate to govern without opposition since the people of New Brunswick chose to elect them with an overwhelming majority.  26 John Gardner observes that the contrast between justificatory defences of wrongdoing and denials of wrongdoing is unclear. 'To say that the defendant committed a criminal wrong, but was justified comes to the same thing, in the end, as saying that she committed no criminal wrong … Regarding some arguments available to criminal defendants it is admittedly unclear, morally as well as legally, whether they are to be regarded as denials of wrongdoing, or rather as concessions of wrongdoing coupled with assertions of offering an all things considered justification understood as a demonstration that we acted for undefeated reasons (162). Instead, our justificatory legal defence speaks only to the reasons for action that we have as legal subjects. Such a defence leaves open the possibility, Lefkowitz says, that, all things considered, we not only acted wrongly but without an all things considered justification. 27 Lefkowitz roots his conception of justificatory defences in an interpretation of Joseph Raz's service conception of law and legal authority, according to which the law's claimed authority is legitimate if, and only if, (1) the person subject to that law better conforms to the reasons that apply to her anyway (to be just, fair, respectful, kind, etc.) if she is guided by the law's directives than if she follows her own reasoning on how to act (the normal justification thesis) and (2) this is a situation in which it's better that she conform to reason than that she decide for herself how to act (the independence condition). 28 Although Lefkowitz highlights a credible distinction between positivistic justificatory defences (i.e., defences that the law chooses to recognise), on the one hand, and all things considered moral justifications, on the other hand, nevertheless his approach has unpalatable implications. Briefly, it implies that, if the law allowed a white person to give a necessity defence for lynching a black person, then a white person would have no duty to follow the law against murder when she lynched a black person. The problem with rooting justificatory defences in legally endorsed 'reasons' is that those 'reasons' might not be reasons at all. My project is about the legal excuses and justifications we should have available to us. Consequently, I do not wish to divorce justificatory defences from justifications and the genuine moral reasons that animate them. When called to account, we want to say correctly that we had no moral duty to follow the law in a given situation, not that we acted according to what the law says are undefeated reasons. When we do say correctly that we had no moral duty to follow the law in a given situation (or that we followed a more pressing moral duty in that situation), we are saying correctly that we genuinely acted for undefeated moral reasons. This approach fits better with Raz's service conception of authority than Lefkowitz's approach does since Raz has in mind the genuine reasons that apply to us, and not legally endorsed 'reasons,' which might not be reasons at all. Although the law claims to have authority over us, it rarely, if ever, meets the tests for a genuine moral authority, according to a service conception.
According to Lefkowitz, a second way that we can attempt to legally justify criminal conduct is to argue that we had a moral right to act as we did. He says that the law has no legitimate authority over us when the independence condition isn't met; that is, the law has no authority over us when it is more important that we decide for ourselves how to act than that we conform to reason. Lefkowitz says that, when we assert in court that we had a moral right to perform a particular law-violating act, we make our case not on the merits of our conduct but by challenging the law's authority over us in relation to that act. Lefkowitz says that: In evaluating her argument, and so considering whether to grant her a justificatory legal defense, the court need not and perhaps should not consider whether her conduct was justified all things considered (i.e., whether she acted for an undefeated reason). Rather, it should ask whether the defendant actually enjoys the right she claims, and if so, whether her conduct fell within the scope of that right. 29 If the answer to both questions is 'yes,' then she has a justificatory defence, Lefkowitz argues.
In reply, justifications track reasons to act, and a moral right is not a reason to act. Moral rights give other people reasons to act; they do not give right-holders reasons to act. If a person owns a stereo, then she has a right to destroy her stereo with a baseball bat. But, her right to do that is not a reason for her to do it. Her right to her stereo is, however, a reason for other people neither to destroy it with a baseball bat nor to stop her from doing so. Similarly, a person's moral right to engage in some conduct that violates the law, such as civil disobedience, is not a reason for her to engage in that conduct. As such, her moral right to civil disobedience cannot form the basis of an assertion that her exercise of that right is a justified (reason-backed) act. 30 Put differently, moral rights of conduct give us a protected sphere of action in which to act wrongly. When we assert a moral right, we are not asserting that we have done no wrong. By contrast, when we offer a justificatory legal defence for criminal conduct, we are, in Lefkowitz's terms, asserting that we did no wrong by failing to treat a particular criminal prohibition as authoritative.
Turning to the demands-of-conviction defence, I root this legal excuse for civil disobedience in, first, a humanistic respect for persons' autonomy as expressive beings (discussed above) and, second, a sensitivity to the psychological costs that persons will suffer if they are required always to privilege the law before their deep convictions. Lefkowitz argues that both my arguments actually support a justificatory demands-ofconviction defence and not an excusatory defence, as I claim. Concerning respect for autonomy, he says: The argument from autonomy implies that when brought before a court of law [the civil disobedient] should offer in her defense not simply the particular conviction that motivated her illegal act but her right as a ''reasoning and feeling being capable of forming deep moral commitments'' to act on her conviction even, within certain limits, when that involves violating the law. 31 In Lefkowitz's view, such a defence is a justificatory defence, as it meets the law's accusation of criminal wrongdoing with the response that she had no duty to treat the criminal norm she violated as proscribing conscientiously motivated communicative acts of disobedience. 29 Lefkowitz (2014), § on The Nature of Justificatory and Excusatory Legal Defenses. 30 Briefly, Lefkowitz seems to embrace the 'legal rights presumption,' i.e., the presumption that, if we have a moral right to something, then there are putative grounds to translate it into a legal right too. But, if the right of conscientious action including civil disobedience were a legal right (which I argue in the book it could not be), then it would be trivial to say that an act of conscientious disobedience was legally justified. It also wouldn't say much in defence of protecting civil disobedience. Legal rights aren't rights in the philosophical sense. They are legal protections, and, in general, what they protect depends on what the enacted positive law protects, not on legitimate claims. 31 Lefkowitz (2014), § on A Demands-of-Conviction Justificatory Defense.
Concerning the psychological costs of always privileging conformity to law over deep conviction, such as self-alienation and akrasia, Lefkowitz cites my observation that, if we value autonomy, we must value the pre-conditions for autonomy, which include psychological integrity and practical reasoning ability. Lefkowitz notes that, … if valuing the conditions for autonomy is part of what is involved in valuing autonomy, and if properly valuing an agent's autonomy requires recognizing her right to conscientious action, then valuing the conditions for autonomy requires recognizing a right to conscientious action (see footnote 31).
In Lefkowitz's view, the assertion of such a right in court equates to giving a justificatory defence.
In reply, although I maintain that the demands-of-conviction defence is an excusatory defence, I grant in the book that one might reasonably think it could have (partially) justificatory elements since the ostensible 'value' that a civil disobedient seeks to protect through her disobedience is not just whatever cause she believes gives her undefeated reasons to act, but also the genuine value of her autonomy, expressive agency, and capacity for deep, persistent commitments (165-166). Together, these elements lend a justificatory shine to what would otherwise be her assertion of mere excusability on the basis of her moral right to act wrongly (or at least her right to act in a way that a truly conscientious agent acknowledges might be wrong).
That said, we must be careful not to misconstrue the value of autonomous choice when what we choose to do is deeply morally wrong. Otherwise the justificatory shine on autonomous choice would be reflected in our conduct when we engage autonomously in murdering and raping people as well as when we engage in suitably constrained conscientious disobedience. In the book, I note that we might hold that the value of autonomy is restricted purely to valuable acts, and so it is not reflected in murders and rapes. Alternatively, we might say that the value of autonomy is not restricted to valuable acts, but to acts that do not interfere with others' autonomy and so, again, it is not reflected in serious wrongs such as murders and rapes. Finally, we might bite the bullet and say that the value of autonomous choice can be present even in serious wrongs such as murders and rapes, but that value is wholly outweighed by the gross disvalue of the acts in question (166). In my view, the jury is out on which option is the most philosophically respectable. My point is simply that, in valuing autonomy, we are not committed to seeing as valuable (that is to say, justified) all acts that are done autonomously.
Lefkowitz notes these observations I make in the book and then presses further, stating that I should explain how 'a conscientiously motivated law-breaker's moral commitments interfere with her responsiveness to the reasons that apply to her, and why, as a concession to human imperfection and frailty, the state ought not to hold her responsible for her defective practical reasoning' (see footnote 31).
In reply, as its name suggests, the demands-of-conviction defence is, by nature, a general defence available to all civil disobedients regardless of their conviction, i.e., regardless of the quality of the reasons that inform their decision to break the law. Being rooted as it is in respect for persons' autonomy and capacity for deep, demanding commitments, this defence does not distinguish between one disobedient's morally credible cause and another disobedient's quirky cause and a third disobedient's morally troubling cause. Thus, the concession to frailty and imperfection that the state makes by excusing disobedients on the grounds of demands-of-conviction is a concession to the fact that we can get it wrong, sometimes badly wrong, in our choice of deepest commitments, but nonetheless we are reasonable to desire to use suitably constrained modes of expression, including constrained breaches of law, to defend our commitments; and, society puts undue pressure on us when it requires us always to privilege the law before our deep commitments.
Lefkowitz offers some highly thoughtful and challenging reflections that bring out the differences between his views and mine on both the moral right to civil disobedience and the legal defence it grounds. These are differences over which he and I are not likely to have a rapid meeting of the minds, and I concede to Lefkowitz that, given the complexity of the conceptual territory of justifications and excuses, there is room for us reasonably to disagree about what kind of defence demands-of-conviction could be.

Conclusion
Constrained by space, I have not addressed all of substantive points made by the four contributors to this symposium on Conscience and Conviction. I hope, though, that I have answered, however imperfectly, the weightiest objections that they raise against my account. Each contributor not only presses me to answer some important challenges, but also identifies places where further investigation of the themes in the book is warranted. Hill's exploration of the conceptual terrain of conviction and conscience shows that more work could be done to situate these two notions within a pluralistic moral framework and to distinguish that framework from monistic moral frameworks. Smith's discussion of rights, burdens, and duties highlights nuances in the relations between protesters and law enforcers, which a fully worked out account of the moral right to civil disobedience would need to accommodate. Weinstock argues for a robust conception of democracy that gives additional strength to the democracy problem, but which, I hope, can ultimately tolerate what Weinstock calls non-remedial civil disobedience. Lefkowitz makes a powerful case for a legal demands-of-conviction defence for civil disobedience, different from my own case. Although I am not inclined to adopt his approach, I would be interested to see it fleshed out in all its details.
It has been a pleasure to write a book on a topic that is beginning to attract interest once again from philosophers. This interest is sparked, no doubt, partly by the frequency of civilly disobedient protests in our time as well as by people's growing use of new forms of disobedience such as digital disobedience and globalised disobedience, which provide a rich new source of conceptual and normative challenges for philosophers to debate, the intellectual fruits of which I look forward to savouring.
Open Access This article is distributed under the terms of the Creative Commons Attribution 4.0 International License (http://creativecommons.org/licenses/by/4.0/), which permits unrestricted use, distribution, and reproduction in any medium, provided you give appropriate credit to the original author(s) and the source, provide a link to the Creative Commons license, and indicate if changes were made.