1. Top of page
  2. Abstract
  3. 1
  4. 2
  5. 3
  6. 4
  7. Conclusion

Acts of civil disobedience, which imply the open violation of a legal directive, often result in the forceful imposition of a choice upon others (e.g. blockades). This is sometimes justifiable, within a democracy, in cases of ‘democratic deficit’, namely, when fundamental rights of an oppressed minority are at stake. In this article, I claim that the use of physical force, in a democracy, may also be justified by the rights of (at least some of) the very people upon whom force is applied. Focusing on the nature of civil disobedience as a ‘form of address’, I argue: (1) using physical force to address others in the democratic arena does not entail infringing upon their status as autonomous agents; (2) using physical force to address others in the democratic arena may contribute to the fulfilment of a positive duty to promote the autonomy of (at least some of) those very people upon whom force is applied. This is not a defence of paternalism: I claim that using force against others, in the democratic arena, may be constitutive of a behaviour that treats others with the respect due to their status as autonomous agents.

In spite of their illegal nature, acts of civil disobedience may be allowed, under democratic regimes, as ‘forms of address’, i.e. as ways to communicate with others in the public arena, by raising a concern with a certain law or policy.1 From this viewpoint, the breach of law represents a ‘speech act’, through which the disobedients aim to communicate their political concerns to fellow citizens.2 Yet, it is also acknowledged that ‘while [civil disobedience] may warn or admonish, it is not itself a threat’, which has led to a general acceptance of non-violence as a necessary condition of civil disobedience.3 This is partly due to the fact that using force against others, in the public arena, would threaten the right of each agent to express a free and uncoerced opinion over the issue at hand.4 The attempt to coerce others shows disrespect for their autonomy, treating them as means to the coercer's ends: democratic citizens ought to employ only persuasive forms of address, so to treat each other with the respect due to autonomous self-legislators.5 That is why force is generally used as one of the benchmarks to distinguish uncivil from civil forms of protest.6

In this article, I seek to challenge this assumption. I will contend that forms of address that employ a degree of force may nonetheless treat others as autonomous agents, and be compatible with the democratic commitment to individual autonomy: I will claim that forcing others to act against their will is compatible with a democratic requirement to treat them as autonomous agents.7 My main goal is to offer a counterargument to the claim that ‘any interference with the civil liberties of others tends to obscure the civilly disobedient quality of one's act’:8 in fact, I will argue that an act of disobedience may be civil even when it forcefully interferes with others’ liberties.

I will defend the following two claims:

  • (1) 
    using physical force to address others in the democratic arena does not entail infringing upon their status as autonomous agents;
  • (2) 
    using physical force to address others in the democratic arena may contribute to the fulfilment of a positive duty to promote the autonomy of (at least some of) those very people upon whom force is applied.

Note that (2) contains the distinctive feature of my argument. It is, in fact, generally accepted that citizens could justifiably employ a degree of force in situations of ‘democratic deficit’, e.g. where the democratic channels are not available due to power imbalances.9 Under these circumstances, the use of force may be justified as part of the attempt to defend one's own rights, or the rights of an oppressed or marginalized minority.10 However, I contend that the use of force in a democracy may sometimes be justified also by the rights of the very citizens upon whom force is applied: that is to say, I will claim that forceful interferences with others’ choices may sometimes be required by the duty to respect them as autonomous agents.

I should stress from the outset that I do not purport to advance the standard paternalistic/utilitarian argument, according to which forcing a subject to act against her will might be permissible if it is in the subject's interest.11 My argument is non-consequentialist, since I will claim that forcing a person to act against her will is, under some circumstances, constitutive of a behaviour that treats that person as an autonomous agent.

This article comprises four sections. In Section 1, I will introduce the following stipulations, to set the stage for my argument in later sections: (a) a conception of autonomy as a person's effective ability to shape the course of her life; (b) an account of respect grounded on the objective worth of the autonomous agent; (c) the notion of a ‘right of autonomy’ and the corresponding negative and positive duties it generates; (d) a distinction between autonomy and freedom.

In Section 2, I present a recent case of civil disobedience. Based on my claims in Section I, I use this case to argue that a forceful interference with a person's choices may infringe upon her freedom without affecting her status as autonomous agent (claim (1) above).

In Section 3, with reference to the same case, I argue that a forceful interference with a person's choices may contribute to the fulfilment of a positive autonomy-based duty to provide her with morally good options from which to choose (claim (2) above).

In Section 4, I consider three objections to my argument.


  1. Top of page
  2. Abstract
  3. 1
  4. 2
  5. 3
  6. 4
  7. Conclusion

The justification of democracy relies on the idea that people are, in general, qualified to govern themselves: for ‘it seems self-evident that people ought not to govern themselves if they are not qualified to do so’.12 This is the reason why, for example, children are excluded from the franchise, based on the assumption they lack the capacity to govern themselves.13 A democratic regime, thus, hinges on the value of individual autonomy, since the former is what best allows for an equal level of self-government or self-legislation among individuals.14 Although the individual is under a (pro tanto) obligation to comply with the electoral outcome even when the results conflict with her particular preference, such preference is deemed to have an influence on the final decision: even though the result cannot satisfy all preferences, it is the result of the confrontation between them.15 From this perspective, the value of a democratic system lies not in the fact that each individual ‘obeys her own will’, but in the fact that each can take part in a process of ‘collective self-legislation’.16 It then becomes clear how the use of force may clash with the basic principles of democracy: it interferes with a person's ability to participate in collective self-legislation, for it seeks to impose choices she may not endorse if uncoerced.17 The attempt to force others fails to acknowledge their status as autonomous agents, who deserve to be engaged with in a fair and uncoerced manner. In fact,

[an a]greement can indeed objectively be obtained by force; but what comes to pass manifestly through outside influence or the use of violence cannot subjectively count as an agreement.18

Focusing on individual autonomy means considering a person's effective ability to be in control of her life: she is autonomous to the extent she can shape the course of her existence, by choosing and pursuing her conception of the good.19 It is this ability to define the course of her own life that makes an autonomous person worthy of respect: a person able to give direction to her existence ought to be respected for this ability.20 Autonomy, therefore, sets a descriptive standard for what is peculiarly human — the capacity to make choices — but also a normative criterion about how autonomous agents ought to be treated. What follows from this is that a person has a right to be respected as an autonomous agent, what I will call a ‘right of autonomy’.21

Now, there are very different ways to show respect for something or someone.22 The attitude of ‘respect’ I focus on is not based on personal inclinations: it does not rely on a subjective evaluation of the person to whom respect is due, but on that person's objective worth or dignity.23 In this sense, we respect a person because of her intrinsic properties, whose worth is independent of our interests and desires. Respecting someone as an autonomous agent, then, contrasts with attitudes such as pursuing, promoting, or using, where the object's value is relative to the subject's interests. This is the gist of Kant's Formula of Humanity, which forbids treating people ‘merely as means’.24

The respect due to an autonomous agent is not grounded on the agent's merit: on the contrary, it involves acknowledging a duty to treat her in a way that is not conditional on her merits or demerits.25 We show respect for someone's dignity not because of what she has achieved, but because of whom she is.26 There may be cases in which respect for autonomy is even contrary to our inclinations, as when we have to respect our enemies, or someone's opinion with which we strongly disagree. A person whose behaviour is morally objectionable still ought to be treated in a way that respects her status as autonomous: even when she acts wrongly, we owe it to her to hold her accountable for her conduct.27

A further important element arising from the objective-groundedness of respecting individual autonomy is that it focuses on disposition rather than outcome: treating a person in accordance with her right of autonomy is not identical with providing for that person's wellbeing. There are cases in which we show our respect for a person's dignity in ways that may inconvenience her.28 This is central to my discussion for, as I have already mentioned, I do not hold that interfering with a person's choices is permissible insofar as it is in that person's interest (i.e. for paternalistic reasons). In fact, I will claim that such interference might display the proper respect for that person, regardless of the consequences. The kind of respect I am focusing on does not aim at someone's welfare, but at her right to be treated in accordance with her status as an autonomous agent.29

As with the concept of ‘respect’ discussed above, the notion of ‘right’ is open to a wide variety of interpretations. A common approach separates ‘negative’ from ‘positive’ rights, and identifies corresponding duties.30 There is, however, a growing dissatisfaction with the negative/positive dichotomy, since many rights have both a positive and a negative side.31 Consider the right to migrate from one country to another, or the right of a linguistic minority to preserve its own language. Both rights demand that others not interfere, but also require the provision of some goods (e.g. transports, in the former case; language schools, in the latter).32 This reveals the implausibility of arguing that leaving a person to her own devices is tantamount with respecting her status as autonomous: someone who is merely free from interferences ‘will have attained a perfectly useless state if she lacks intelligence, skills, knowledge, and emotional health’.33

Therefore, I argue that respecting a person's right of autonomy involves more than mere non-interference: though sharing the same capacity (i.e. the same dignity as autonomous agents), individuals differ as to whether they are in the actual condition to make autonomous choices.34 A person kept prisoner by another would not be able to exercise her capacities, due to the situation she finds herself in.35 Similarly, someone who is denied basic education, while free from interferences, will not be able to fully exercise her capacity to make choices. Hence, I hold that the right of autonomy bestows both negative and positive duties on others.36 We acknowledge a person's dignity not simply by refraining from interfering with her choices: we should also see to it that she is in the condition to make choices.37 If we ought to respect a person's right to shape the course of her life, then we should also take some positive steps to promote her exercise of that right.38

Evidence for this claim is provided by the fact that autonomy has a social dimension: a person needs other people in order to be autonomous.39 When we exercise our capacity to make autonomous choices, we apply standards that are not entirely determined by ourselves, but are partly the product of the context we have grown into.40 We have not determined these standard ex nihilo: we have acquired them, at least in part, from others’ examples and teachings.41 From this standpoint, autonomy requires both a degree of distance from and of proximity to others, that is, both independence and support: it is something that must be learned, and such learning requires a background of shared values against which autonomous choices may be made.42 While being autonomous implies achieving a degree of independence of choice, the latter can be realised only from within a life shared with others. An effective ability to make choices involves the power to revise or reject our own projects according to criteria determined, in part, by the context we grow up in.

I now seek to introduce a distinction that will be central to the following discussion. For the purposes of this argument, I will distinguish the notion of ‘autonomy’ from that of ‘freedom’, holding the latter to concern a realm of human agency narrower than that pertaining to autonomy. To borrow Gerald Dworkin's terminology, I take autonomy as a ‘global’ concept, and freedom as a ‘local’ one: that is, while a person can be said to be ‘free’ at a specific moment in time, the assessment of her autonomy involves evaluating a substantial section of her life.43 The latter involves considering whether the person is actually in control of her life, i.e. whether she is able to pursue the projects and values she deems central to her conception of the good.

The role played by autonomy in a person's life does in fact connect to her identification with her projects, values, goals and desires. There is a peculiar difference between, for example, the choice to have coffee rather than tea, and the choice to support a certain policy or political party: the latter mirrors, to some extent, the choice to be a particular kind of person, who endorses a certain conception of the good (in a way that the choice to drink coffee over tea would not normally do).44 As I will argue in the next section, being autonomous involves more than just being free: as a character ideal of persons who are in control of their lives, autonomy requires a certain degree of freedom, yet freedom by itself is not sufficient for making autonomous choices.45 This will also reveal that, in some cases, we may interfere with a person's choices without infringing on her autonomy.


  1. Top of page
  2. Abstract
  3. 1
  4. 2
  5. 3
  6. 4
  7. Conclusion

In this section, I will claim that forcing others to act against their own will may be compatible with a negative duty not to interfere with their autonomous choices. In Section 3, I will then argue that this kind of behaviour may also contribute to the fulfilment of a positive duty to support their autonomous choices.

Consider this case of civil disobedience:

In the early hours of the 8th of December 2008, fifty-seven activists cut through the security fences surrounding the runway at Stansted Airport, UK. They assembled fencing panels on the runway, and hooked themselves to the panels with chains and D-locks, while wearing high-visibility vests with the inscription ‘Please Do Something’ on the back. When police arrived, the activists did not try to resist arrest. Their act of protest deliberately caused the cancellation of several flights, and a serious disruption to passengers due to travel on that day. The protesters aimed at highlighting the environmental risks connected to the UK Government's plan to build a third runway at Stansted Airport.46

Through this action, the activists intentionally forced other people to act against their own will. Passengers had to withdraw (at least provisionally) their travelling plans, facing a serious inconvenience.47 This protest was intended to be an act of civil disobedience, i.e. a way to address others in the public arena. However, notwithstanding the action's communicative aims (i.e. to offer an objection to a proposed policy), the activists used a degree of force: they sought to convey their message by imposing a choice (i.e. staying at the airport) on others, disregarding the latter's will. The presence of force seems to put the action at odds with the democratic commitment to autonomy, since it may reveal a failure to show the appropriate respect for the status of autonomous agents. Contrary to this view, I hold that the Stansted case illustrates a form of address that is compatible with the requirement to respect others as autonomous agents (who have a right to make uncoerced choices).

To begin with, we should look at the case of the Stansted protest from the perspective of the autonomy/freedom distinction presented in Section 1 above. In light of this distinction, I claim that, although the protesters who occupied the airport applied a degree of force on other people, that force was used to (temporarily) constrain only others’ freedom to move: yet, the protest did not infringe upon others’ autonomy, for it did not aim to coerce others into opposing the planned expansion of the airport.

Remember that the distinction between ‘freedom’ and ‘autonomy’ is based on the different realms of human agency to which these two notions apply. Autonomous choices, I have claimed, concern the direction a person seeks to give to her own life. Some choices of ours do not say much about the kind of persons we are, although we are free to make them (e.g. boarding on a certain flight). Other choices do: choosing to support a certain policy or government says something about the kind of persons we are, what values we subscribe to, what conception of society we endorse.48 This is a choice for which we bear responsibility, since it shows our commitment to a certain idea of the good: and this is why interfering with this kind of choice, by forcing us to do otherwise, may seriously infringe on our status as autonomous agents.

What follows from this is that force employed against a person's freedom does not necessarily infringe also on her right of autonomy: by interfering with your choice to board on a certain flight, I do impinge on your freedom, though not necessarily also on your autonomy.49 My claim is that, in the Stansted case, activists did infringe only on others’ freedom, not on their autonomy to choose whether to support the planned airport expansion.

In order to spell out my view in more detail, I should first clarify how the Stansted case differs from one (call it ‘chair case’) in which A ties B to a chair, and does not let B go until B accepts to engage in a discussion with A concerning the planned airport expansion.50 In the chair case, B is indeed coerced into engaging in a discussion.51 However, the Stansted case differs in two main respects from the chair one. First, the protesters did not introduce threats: their message was not ‘we will keep you here until you listen to us’. Rather, passengers were bluntly forced to stay there and wait for the police to remove the protesters. This is relevant here, for it shows a difference based on duration. The Stansted protest occurred over a very limited segment of time: it was not designed to keep people stranded unless they surrendered to the protesters’ desires. The latter behaviour would certainly be closer to kidnapping than to civil disobedience.

A second, and crucial difference between the ‘Stansted’ and the ‘chair’ cases concerns what the subjects are forced to do. In the chair case, A ties B to a chair in order to force B to have a discussion; in the Stansted case, this does not occur. There is no way, strictly speaking, for the protesters to actually force others at the airport to engage in a discussion: their action is a form of address, which nonetheless can still be ignored by the addressee.52 Although they do employ a degree of force, they do not go as far as coercing the passengers to pay attention to their message. The latter can still choose to ignore what the protesters are saying, without being forced to make any choice as to what policy to support: while the protest infringes upon their freedom, it does not impact on their autonomy. This shows that force is not incompatible with a democratic regime: there are cases in which a degree of force may be part of a (legitimate) attempt to get the hearer's full attention, yet the hearer can still make her own, uncoerced choice about the issue at hand, including the choice to ignore the speaker's message.53 When force is employed not as a means of intimidation, but to ensure that the hearer could grasp the substance of the speaker's view, the act still abides by the (negative) duty not to interfere with others’ autonomous choice-making.

Hence, I argue, the presence of force in civil disobedience does not entail coercion of individual autonomy. Rather, actions like the Stansted protest may represent instances of what Hannah Arendt labelled ‘communicative power’, that is, ‘the capacity to act in concert’: by engaging each other in a rational exchange, citizens collectively exercise their power as law-makers (i.e. as autonomous agents).54 In Arendt's words, ‘[p]ower is never the property of an individual; it belongs to a group and it remains in existence only so long as the group keeps together’.55 Provided the act remains communicative in nature, and that it abides by a commitment to persuasion, it represents a way to empower the hearers by calling for an answer, that is, by appealing to their status as autonomous agents. Addressing others in this way acknowledges their right to take part in the process of collective self-legislation.

We should, then, be wary of criticizing an action as ‘undemocratic’ merely because it forces people to act against their will. Some acts that interfere with the behaviour of others might temporarily impinge on their freedom, without affecting their status as autonomous participants in the democratic discourse. An excessive emphasis over the idea of ‘rationality’ as calm and self-restrained behaviour might lead to the exclusion, from the deliberative arena, of genuine forms of communication.56 Attempting to have others listen does not mean forcing them to choose: it is a call for a dialogue, not an order to accept one's own view. A temporary infringement of a person's freedom does not entail the coercion of her autonomy.

However, the breach of freedom still requires a justification: I am not suggesting that it is fair to infringe on a person's freedom as long as her autonomy remains intact. That is, I still need to explain what justified breaching others’ freedom in the Stansted case. I address this issue in Section 3.


  1. Top of page
  2. Abstract
  3. 1
  4. 2
  5. 3
  6. 4
  7. Conclusion

There is an important element of self-identification in the condition of autonomy: an autonomous person is able to see the projects and commitments in her life as her own, not as imposed on her by someone else. As Frankfurt and Dworkin have argued, someone who has a certain desire might also resent having it (as in the case of addiction), and this would raise doubts as to her self-identification with that first desire.57

From this standpoint, an autonomous choice expresses the agent's self-identification not only with her own desires, but also with the motivations behind those desires. In addition to saying ‘I want to X’, an autonomous person is in the condition of saying ‘I want to want to X’.58 Hence, if a person regrets having a certain motivation. i.e. would like to be motivated by different influences, the motivation, though effective, might not be viewed as her own. An autonomous person makes ‘authentic’ choices, i.e. choices involving a degree of self-identification also with her higher-order desires.59

A choice that is not authentic in the sense just explained might not qualify as ‘autonomous’, for it would lack the agent's endorsement.60 If this is the case, then we might interfere with that choice without necessarily infringing on the agent's autonomy: rather, the latter might be promoted, if the interference was to enhance the agent's self-identification with her own choices and desires. Having recognised the source of her desire, the agent could then either wish she were not motivated in this way, rejecting that desire as not her own, or reaffirm it.61 In this case, the interference could support, rather than hinder, that person's exercise of her capacity as autonomous agent.

I claim we should look at the interference with others’ freedom in the Stansted case from this same perspective, that is, as aiming to promote others’ autonomy. One of the protesters’ goals was to provide other citizens, including the very ones forced to remain at the airport, with information about the risks of further airport expansion in the face of growing evidence of human responsibility for climate change. My suggestion is that that action may be interpreted as addressing others with the question whether they would want to support a policy that could contribute to endangering the ecosystem. Through their speech act, the protesters were asking their fellow citizens to take a stand on an important issue, in light of the fellow citizens’ own values: that is to say, in light of the kind of persons fellow citizens have chosen to be. As I explain in more detail below, by doing so the protesters were addressing others with the respect due to the their status as autonomous agents.

While, as I showed in the previous section, the Stansted protest complied with a negative duty not to coerce others’ autonomous choices (since, I argued, it targeted others’ freedom only), I contend that it also contributed to the fulfilment of a positive duty to support the exercise of others’ capacities qua autonomous agents, fostering their self-identification with their own choices. This can be grasped by focusing on one of the aims of political protest, i.e. the provision of information about a particular issue or policy: I claim that actions like the Stansted protest seek, among other things, to offer valuable options for others to choose from.

This takes us back to the social dimension of autonomy, which I discussed in Section 1: we rely on others, to a certain degree, to develop and exercise our capacities for autonomous living. One way in which the surrounding social environment supports the exercise of our autonomy is by providing us with options: we cannot choose what kind of person to be, what kind of life to have, what kind of values to subscribe to, if we do not have alternatives available. The provision of valuable options, therefore, represents one way towards fulfilling a positive duty to promote others’ autonomy.62

To grasp this point, let us consider the difference between choosing option A out of two options and out of ten.63 In the second case, the choice bestows a higher degree of responsibility on the agent, since it results from a reflection over a much larger range of possibilities. By choosing A in the second case, the agent also chooses not-B, not-C, … , not-J: in this sense, she determines ‘what she does not do as well as what she does’, and this enhances her responsibility, and accountability, for her choice.64 In fact, her choosing A out of ten options says more about what kind of person she is than her choosing it out of two options only: it reveals a richer identification with her own desires, goals and values. To every ‘choosing to’ there correspond a number of ‘choosing not to’, so that ‘rather than deciding to create oneself, one is choosing to reject some of what one is in the light of other commitments or values.’65

Note here the emphasis is on quality rather than quantity: providing a number of very similar options does not enhance the agent's responsibility for choosing one over the other. By the same token, a choice between various morally repugnant actions does not enhance one's autonomous condition: to promote one's autonomy, we must provide morally valuable options.66

Let us now return to the Stansted case. I contend that the use of force was part of the attempt to provide fellow citizens with valuable options from which to choose. By highlighting the environmental risks caused by airplane traffic, the protesters sought to enhance others’ responsibility; their message raised a question not only about what policy fellow citizens would support, but also about what kind of persons they would choose to be. The protesters appealed to the status of agents that see their desires and motivations as authentically their own. As I already explained, those forced to remain at the airport were not being coerced, for they were still able to choose whether or not to join the disobedients’ plea. However, after acknowledging the protesters’ message, their choice would have had a richer meaning: it would now be a choice made against a set of options that contains ‘caring for the environment’. The non-disobeying citizens would assess their own values according to a ranking that also comprises environmental ideals. After facing the disobedients’ message, their choice will say more about what kind of persons they have chosen to be, i.e. what values they endorse: they might support economic advantage (if any) over environmental concerns, by rejecting the disobedients’ plea, or vice versa, by deciding to oppose the government's plan for airport expansion. Either way, they will be held accountable, to a higher degree, for the choice they have made — and treated accordingly (i.e. blamed or praised). In both cases, their self-identification is not diminished, but is, if anything, promoted by the act of protest.

Thus, although the protesters used force against others, they respected others’ autonomy, since people were left uncoerced with regard to what policy to support. Furthermore, I have argued that forcing others to act against their will may be part of the attempt to enhance their power to ‘define their nature, give meaning and coherence to their lives, and take responsibility for the kind of person they are’.67 Acts of civil disobedience that, as in the Stansted case, employ a degree of force, may fulfil both negative and positive obligations towards others’ autonomy.


  1. Top of page
  2. Abstract
  3. 1
  4. 2
  5. 3
  6. 4
  7. Conclusion

A first objection to my argument is that it would not apply to those citizens who have already committed themselves to environmental ideals. Among the passengers stranded at the airport, there were surely citizens who were already sympathetic to the disobedients’ cause; these citizens were not being provided with a new option from which to choose, for they had already chosen it! Similarly, others may have already chosen to support the airport's expansion vis-à-vis the environmental risk, deeming economy more important than nature: here as well, the ‘environmental’ option was already available (and yet, was not chosen). In both cases, the subjects had already chosen ‘what kind of person to be’ with reference to the options the protesters were presenting. Thus, my argument could only apply to a specific group of citizens, those who, for lack of time, resources, or interest, did not possess sufficient knowledge about the issue at hand. Only in their case we could say their autonomy may have been promoted by the protest. In the case of those who had already ‘chosen’ which option (economy versus environment) to support, the protest could not enhance self-identification.

This is why I claimed that the use of force in civil disobedience could promote the autonomy of at least some of those upon whom force is applied. Also, while my argument for a positive contribution to others’ autonomy through the use of force may apply only to some cases, my previous claim that these actions infringe upon no one's autonomy, do not coerce others’ choices, and are compatible with democratic principles still holds. As long as others’ right to accept or reject the substance of the message is preserved, the action should not be condemned as incompatible with the democratic commitment to individual autonomy.

A second objection concerns my claim that the activists at Stansted airport, through their illegal protest, were fulfilling their duties as democratic citizens.68 While I have explained this claim in light of a duty to respect others as autonomous agents, I seem to have ignored that citizens are, first and foremost, under a moral obligation to obey the law. How can an act of defiance of the law be justifiable as the fulfilment of one's own political obligations?69

The account of political obligation defended here relies on the assumption that citizens’ political duties are not exclusively fulfilled through acts of obedience.70 In different contexts, a person may discharge her political duties by, for example, resorting to illegal conduct. When this is the case, the act of disobedience can be seen, under the circumstances, as not merely permissible, but required by the agent's own political obligations.71 Thus, I challenge the assumption that disobedience always needs a justification: there is nothing ‘wrong’ per se with disobedience, in the same way in which there is nothing ‘right’ per se with obedience.72 The focus on the agent's disposition and motivation, rather than on the external features of her act, can properly capture the democratic nature of her behaviour.73

This does not mean that the protesters are exempted from facing the legal consequences of their action. One of the central assumptions of civil disobedience is the agents’ willingness to undergo the punishment attached to the law-breaking behaviour. There might be a series of mitigating circumstances that could even lead to the protesters’ acquittal: but still, the protesters face a requirement to be willing to expose themselves to the legal consequences of their action.74

A third objection concerns the amount of force I am willing to allow. Although I have claimed that a person may breach another's freedom without also impinging on her autonomy, I have not offered any criterion as to how far the use of force can go. This is consistent, however, with the context-sensitive account of political obligation I have just sketched. In some cases, we might have to temporarily lock someone up in order to have her attention, if we think the issue warrants that kind of action: in other situations, we might have to do even something more. Friendship may offer a good analogy here: if I thought a close friend of mine was going to do something very dangerous or stupid, I might have to slap her on the face in order to have her listen to me, if she were unwilling to heed my concerns. If after being forced to listen to me, she confirmed her intention to go ahead with her plan, I might have to step back and let her act as she wants. But was I to let her go without engaging her to this extreme, I might fail to live up to my obligations as friend: I owe it to her, as my friend, to try to persuade her that she is wrong. Similarly, in a democratic society, we may owe it to our fellow citizens to try to have them listen to us, by all available means compatible with respecting their status.

That said, there are strong reasons against the use of force, and by no means do I want to ignore them. First, forceful behaviours could destabilize the political life of a society, by generating feelings of resentment in those whose freedom is impinged upon.75 Second, forceful behaviours might encourage the use of force in cases where it would be inappropriate or unwarranted.76 Third, force might cause harm to others.77 For reasons like these, the choice to employ force to address others must be carefully scrutinized by the agents, who must be able to consider whether the effects of their action warrant a different behaviour.


  1. Top of page
  2. Abstract
  3. 1
  4. 2
  5. 3
  6. 4
  7. Conclusion

Democracy relies on the status of each participant as autonomous, a ‘self-legislator’, with a right to make uncoerced choices. Coercion is incompatible with a person's right of autonomy, and has no place within a democratic society.78

In this article, I have tried to shed light on the implications of this view, suggesting that forms of address that force others to act against their will do not always defy democratic values. I have focused on civil disobedience, whose justifiability in a democracy is usually tied to a requirement of non-violence. I have argued that some cases of forceful protest target others’ freedom, without infringing on their autonomy. I have further argued that force may also fulfil positive duties to promote others’ autonomous choice-making, by fostering self-identification or authenticity. The gist of this article lies in the emphasis on agents’ political obligations in democratic societies. While, I have argued, obedience to the law is not the only way for us to discharge our duties qua citizens, I aimed to stress that ‘citizenship is a call for action’, which requires agents to engage with their particular context in order to identify the appropriate conduct.79 The latter, under some circumstances, may involve a departure from ‘conventional’ forms of participation, in favour of actions that, while being ‘forceful’, remain loyal to democratic ideals.

Civil disobedience, even when it employs a degree of force, may still qualify as a democratic form of address. We should, therefore, be wary of dismissing actions like the Stansted protest as uncivil: there might be cases in which these forms of behaviour are required by the very principles underlying a democracy. The protesters, in some cases, are just doing their duty as citizens.80

  1. 1

    Civil disobedience involves the ‘politically motivated breach of law designed either to contribute directly to a change of a law or of a public policy or to express one's protest against, and dissociation from, a law or a public policy’: Joseph Raz, The Authority of Law (Oxford: Clarendon Press, 1979), p. 263. On civil disobedience as a form of address, see John Rawls, A Theory of Justice (Oxford: Oxford University Press, 1999), p. 321; Andrew Sabl, ‘Looking forward to justice: Rawlsian civil disobedience and its Non-Rawlsian lessons’, Journal of Political Philosophy 9,3 (2001): 307330.

  2. 2

    B. Smart, ‘Defining civil disobedience’ in H. A. Bedau (ed.) Civil Disobedience in Focus (London: Routledge, 1991), pp. 198211; Kimberley Brownlee, ‘Features of a paradigm case of civil disobedience’, Res Publica: A Journal of Legal and Social Philosophy 10 (2004): 337351.

  3. 3

    Rawls op. cit., p. 321. In this article, I use the terms ‘violence’ and ‘force’ as synonyms, and am concerned only with cases where they are employed to impose a choice on others.

  4. 4

    For the purposes of my argument, I hold that coercion ‘occurs when one person threatens to visit some evil or unwanted consequences on another unless that other does or refrains from doing some act in accordance with the coercer's demand’: B. Leiser, ‘On coercion’ in D. A. Reidy & W. J. Riker (eds) Coercion and the State (New York: Kluwer/Springer, 2008), p. 31

  5. 5

    Here I focus on citizens coercing each other in the democratic arena. I do not challenge the right of the state to coerce citizens into obedience. As I explain in Section 4, my view is not anarchist.

  6. 6

    On ‘civility’, see Cheshire Calhoun, ‘The virtue of civility’, Philosophy and Public Affairs 29, 3 (Summer 2000): 251275.

  7. 7

    As I explain in more detail in Section 4, I seek to justify only the use of ‘a degree of force’, since a form of address involving too much force, e.g. physically harming someone, may not be justifiable as civil disobedience.

  8. 8

    Rawls op. cit., p. 321.

  9. 9

    See J. Morreall, ‘The justification of violent civil disobedience’, in Bedau op. cit., pp. 130143; Rawls op. cit., pp. 309, 319; Sabl op. cit., p. 312; Brownlee op. cit.; William Smith, ‘Democracy, deliberation and disobedience’, Res Publica: A Journal of Legal and Social Philosophy 10 (2004): 353377; William Smith, ‘Civil disobedience and the public sphere: Reflections on Habermas’, Journal of Political Philosophy 19,2 (June 2010): 145166.

  10. 10

    For example, the temporary infringement of someone's right to move freely, through the illegal occupation of a public space, might be justified as part of a protest that asks for the suppression of a law that seriously impinges on someone else's human rights. Ronald Dworkin, Taking Rights Seriously (Cambridge, MA: Harvard University Press, 1977), pp. 206222. See also Morreall op. cit., and Sabl op. cit.

  11. 11

    On paternalism, see Seana Valentine Shiffrin, ‘Paternalism, unconscionability doctrine, and accommodation’, Philosophy and Public Affairs 29 (2000): 205250.

  12. 12

    Robert Dahl, Democracy and its Critics (New Haven, CT: Yale University Press, 1989), p. 95.

  13. 13

    For a recent challenge to this assumption, see Joanne Lau, ‘Two arguments for child enfranchisement’, Political Studies 60,4 (December 2012): pp. 860876.

  14. 14

    See R. Eckersley, ‘Greening liberal democracy: The rights discourse revisited’ in B. Doherty (ed.) Democracy and Green Political Thought: Sustainability, Rights and Citizenship (London: Routledge, 1996), pp. 212236

  15. 15

    Bernard Manin, ‘On legitimacy and political deliberation’, Political Theory 15,3 (1987): 338368, at p. 359. On the notion of pro tanto obligations, see Shelly Kagan, The Limits of Morality (Oxford: Oxford University Press, 2002), p. 17

  16. 16

    Christian Rostbøll, Deliberative Freedom: Deliberative Democracy and Critical Theory (Albany, NY: SUNY Press, 2008), pp. 104105, 210.

  17. 17

    On the notion of ‘endorsement’, see Harry Frankfurt, The Importance of What We Care About (Cambridge: Cambridge University Press, 1988) and Necessity, Volition and Love (Cambridge: Cambridge University Press, 1999).

  18. 18

    M. Cooke (ed.), On the Pragmatics of Communication: Jurgen Habermas (Cambridge, MA: MIT Press, 1998), p. 120.

  19. 19

    Steven Wall, Liberalism, Perfectionism and Restraint (Cambridge: Cambridge University Press, 1998), p. 128. As I say below, this means that a person could be more or less autonomous than another.

  20. 20

    See Marina Oshana, Personal Autonomy in Society (Aldershot: Ashgate, 2006), p. vii; Marilyn Friedman, Autonomy, Gender, Politics (New York: Oxford University Press, 2003), pp. 7374; Joseph Raz, Value, Respect and Attachment (Cambridge: Cambridge University Press, 2001), p. 158.

  21. 21

    Thomas Hill, Jr, Autonomy and Self-Respect, (Cambridge: Cambridge University Press, 1991), p. 48; Richard Dagger, Civic Virtues: Rights, Citizenship and Republican Liberalism (Oxford: Oxford University Press, 1997), pp. 3032.

  22. 22

    See, for example, Joel Feinberg, ‘Some conjectures on the concept of respect’, Journal of Social Philosophy 4 (1975): 13; Stephen D. Hudson, ‘The nature of respect’, Social Theory and Practice 6,1 (1980): 6990; Carl F. Cranor, ‘On respecting human beings as persons’, Journal of Value Inquiry 17,2 (1983): 103117; Robin S. Dillon, ‘Respect for persons, identity and information technology’, Ethics and Information Technology 12,1 (March 2010): 1728.

  23. 23

    B. Herman (ed.), John Rawls: Lectures on the History of Moral Philosophy (Cambridge, MA: Harvard University Press, 2000), p. 153.

  24. 24

    Allen Wood, Kant's Ethical Thought (Cambridge: Cambridge University Press, 1999), p. 18.

  25. 25

    See Stephen Darwall's distinction between ‘appraisal’ and ‘recognition’ respect, in The Second-Person Standpoint: Morality, Respect and Accountability (Cambridge, MA: Harvard University Press, 2006), pp. 122130.

  26. 26

    Darwall op. cit., p. 123.

  27. 27

    ‘[A] sane offender may claim a right to be punished rather than be subjected to some other […] preventive treatment: to punish her is to treat her still as a rational agent’: Antony Duff, Trials and Punishments (Cambridge: Cambridge University Press, 1986), p. 186. See also Peter Strawson, Freedom and Resentment and Other Essays (Oxford: Routledge, 2008), pp. 128.

  28. 28

    See 27.

  29. 29

    See Darwall's distinction between respect and ‘care’, op. cit., pp.126127.

  30. 30

    For an example, see David Miller, Social Justice (Oxford: Oxford University Press, 2002), p. 60.

  31. 31

    Henry Shue, Basic Rights: Subsistence, Affluence, and U.S. Foreign Policy (Princeton, NJ: Princeton University Press, 1980); Stephen Holmes & Cass Sunstein, The Costs of Rights (New York: Norton, 1999), chapter 1; Joseph Raz, The Morality of Freedom (Oxford: Clarendon, 1986); Amartya Sen, Development as Freedom (New York: Knopf, 1999); Sandra Fredman, Human Rights Transformed: Positive Rights and Positive Duties (Oxford: Oxford University Press, 2008); Kai Möller, ‘Two conceptions of positive liberty: Towards an autonomy-based theory of constitutional rights’, Oxford Journal of Legal Studies 29,4 (2009): 757786.

  32. 32

    See Dagger op. cit., p. 15. See also 42.

  33. 33

    Bernard Berofsky, Liberation from Self: A Theory of Personal Autonomy (Cambridge: Cambridge University Press, 1995), p. 15.

  34. 34

    On autonomy as ‘capacity’ versus ‘condition’, see Joel Feinberg, Harm to Self: The Moral Limits of the Criminal Law (Oxford: Oxford University Press, 1986), p. 31.

  35. 35

    Raz op. cit., 1986, pp. 373374.

  36. 36

    Such a view is defended, among others, in Robert Young, Autonomy: Beyond Negative and Positive Liberty (New York: St. Martin's Press, 1986); Marina Oshana, ‘Personal autonomy and society’, The Journal of Social Philosophy 29 (1998): 81102; Rawls op. cit.; Thomas W. Pogge, Realizing Rawls (Ithaca, NY: Cornell University Press, 1989); Jurgen Habermas, Between Facts and Norms (Cambridge, MA: MIT Press, 1996).

  37. 37

    Tom Beauchamp & John Childress, Principles of Biomedical Ethics (Oxford: Oxford University Press, 2001), p. 63; Alan Gewirth, The Community of Rights (Chicago, IL: The University of Chicago Press, 1996), p. 164.

  38. 38

    This is compatible with arguing that negative duties are more stringent than positive ones. The positive duty to respect individual autonomy could be conceived as a Samaritan duty, whereby one ought to take positive action to help others, provided this can be done at ‘no unreasonable cost’. See C. H. Wellman, ‘Samaritanism and the duty to obey the law’ in C. H. Wellman & A. J. Simmons (eds) Is There a Duty to Obey The Law? (Cambridge: Cambridge University Press, 2005), p. 3.

  39. 39

    ‘[T]he awareness of ourselves as selves capable of choice is something that others teach us, wittingly or not’: Dagger op cit., p. 39.

  40. 40

    On the relational dimension of autonomy, see J. Anderson & A. Honneth, ‘Autonomy, vulnerability, recognition and justice’ in J. Christman & J. Anderson (eds) Autonomy and the Challenges to Liberalism: New Essays (Cambridge: Cambridge University Press, 2005), pp. 127149.

  41. 41

    Surely others can also threaten our autonomy: conformism and fear of social stigmata may lead to choices we would not endorse in the absence of social pressure.

  42. 42

    Susan Mendus, Toleration and the Limits of Liberalism (London: McMillan, 1989), pp. 9697.

  43. 43

    Gerald Dworkin, The Theory and Practice of Autonomy (Cambridge: Cambridge University Press, 1988), pp. 1516. My focus here is on ‘negative freedom’ only. Assessing a person's ‘republican freedom’, i.e. ‘freedom from domination’, may involve a person's situation over a certain period of time. See Philip Pettit, Republicanism: A Theory of Freedom and Government (Oxford: Clarendon Press, 1999), pp. 17126. Thanks to an anonymous referee for pointing this out.

  44. 44

    See Geoffrey Brennan & Loren Lomasky, Democracy and Decision: the Pure Theory of Electoral Preference (Cambridge: Cambridge University Press, 1993), p. 186: ‘To cast a Klan ballot is to identify oneself in a morally significant way with the racist policies that the organization espouses’.

  45. 45

    Young op. cit., p. 8; Lawrence Haworth, Autonomy: an Essay in Philosophical Psychology and Ethics (New Haven, CT: Yale University Press, 1986), pp. 166167. Only ‘a certain degree of freedom’ is necessary (though not sufficient) for autonomy, since I argue below that some (limited) infringements on a person's freedom may not affect her autonomy.

  46. 46

    Jenny Percival, ‘Climate activists held after Stansted runway protest’, The Guardian 8 December (2008).

  47. 47

    Of course, the protest did not inconvenience only passengers, but also airport staff, neighbouring businesses, etc.

  48. 48

    Kai Möller stresses this very same point in op. cit., p. 772.

  49. 49

    Robert Nozick, Socratic Puzzles (Cambridge, MA: Harvard University Press, 1997), p. 15: ‘If I lure you into an escape-proof room in New York and leave you imprisoned there, I do not coerce you into not going to Chicago though I make you unfree to do so.’

  50. 50

    Thanks to an anonymous referee for pressing me to clarify this distinction.

  51. 51

    ‘Some people may want to make their decisions impulsively, without rational deliberation: insisting that they hear rational argument […] is paternalism’: G. Dworkin op. cit., p. 123.

  52. 52

    The same cannot be said of the chair case, since the example is designed in a way such that A will not let go of B unless B accepts to engage in a discussion with A.

  53. 53

    Habermas introduces the notion of ‘communicative freedom’, namely the freedom to say ‘no’ to a certain validity claim. This, I argue, was not infringed upon by the protesters at Stansted airport. See K. Gunther, ‘Communicative freedom, communicative power, and jurisgenesis’ in M. Rosenfield & A. Arato (eds) Habermas on Law and Democracy: Critical Exchanges (Berkeley, CA: University of California Press, 1998), p. 237.

  54. 54

    Hannah Arendt, On Violence (New York: Harcourt Brace Jovanovich, 1970), p. 50.

  55. 55


  56. 56

    Iris Marion Young, Inclusion and Democracy (Oxford: Oxford University Press, 2000), pp. 3751.

  57. 57

    Frankfurt op. cit.; G. Dworkin op. cit.; G. Dworkin, ‘The concept of autonomy’ in J. Christman (ed.) The Inner Citadel: Essays on Individual Autonomy (Oxford: Oxford University Press, 1989), pp. 5462.

  58. 58

    See Robert Olick, Taking Advice Directives Seriously: Prospective Autonomy and Decisions Near the End of Life (Washington, DC: Georgetown University Press, 2001) pp. 86 ff.

  59. 59

    G. Dworkin op. cit., 1988, pp. 1820.

  60. 60

    ‘[W]hat makes a preference autonomous is that it has survived a certain process’: Christian Rostbøll, ‘Preferences and paternalism: On freedom and deliberative democracy’, Political Theory 33,3 (2005): 377.

  61. 61

    G. Dworkin op. cit. 1989, p. 59.

  62. 62

    Raz op. cit., 1986, p. 375, 408.

  63. 63

    Thomas Hurka, Perfectionism (Oxford: Oxford University Press, 1993), pp. 149150.

  64. 64

    Ibid., p. 150.

  65. 65

    Gerald Gaus, Social Philosophy (New York: Sharpe, 1999), p. 130.

  66. 66

    This is not to say that someone who chooses evil over good is not acting autonomously. The point is that we do not have an obligation to provide her with the evil options. Compare Raz op. cit., pp. 411412.

  67. 67

    G. Dworkin op. cit., p. 20.

  68. 68

    Here I use the terms ‘duty’ and ‘obligation’ interchangeably.

  69. 69

    Note that the answer here cannot be framed in terms of just versus unjust laws. Although the obligation to obey the law is generally taken to apply only to the not-unreasonably-unjust legal directives (i.e. as a pro tanto one), the Stansted protest was a case of indirect disobedience: the protesters were violating a law they accepted as legitimate (regulating airport traffic) to protest a different law (about airport expansion). On indirect civil disobedience, see William Edmundson, Three Anarchical Fallacies: An Essay on Political Authority (Cambridge: Cambridge University Press, 1998), p. 57.

  70. 70

    A view defended at length in Bhikhu Parekh, ‘A misconceived discourse on political obligation’, Political Studies 41,2 (1993): 236251.

  71. 71

    Thus, it would have been wrong for the Stansted protesters, under the circumstances, not to resort to civil disobedience: acquiescence with a situation deemed unjust or, on the other hand, the appeal to more violent means to coerce the autonomy of others, would have been wrong.

  72. 72

    John Simmons, Justification and Legitimacy: Essays on Rights and Obligations (Cambridge: Cambridge University Press, 2003), pp. 117118; Michael Walzer, Obligations: Essays on Disobedience, War and Citizenship (Cambridge, MA: Harvard University Press, 1970), p. 16.

  73. 73

    For similar suggestions, see William Edmundson, ‘The virtue of law abidance’, Philosopher's Imprint 6,4 (December 2007): 121; Parekh op. cit.

  74. 74

    P. Moraro, ‘A dilemma for the civil disobedient: Pleading guilty or not guilty in the courtroom’, in C. Michelon , G. Clunie , C. McCorkindale & H. Psarras (eds), The Public in Law: Representation of the Political in Legal Discourse (Aldershot: Ashgate, 2012), pp. 99113; Jennet Kirkpatrick, Uncivil Disobedience: Studies in Violence and Democratic Politics (Princeton, NJ: Princeton University Press, 2008), chapter 1.

  75. 75

    Daniel Markovits, ‘Democratic disobedience’, Yale Law Journal 114,8 (2005): 18971952, at p. 1936.

  76. 76

    Raz op. cit., 1979, p. 267.

  77. 77

    The life of someone due to fly for heart surgery from Stansted airport, on the day of the protest, would be put at risk by the protesters’ action.

  78. 78

    See 5.

  79. 79

    ‘[F]or the end [of political science] is action, not knowledge’: Aristotle, Nicomachean Ethics I, 4, 1095a5. See also Benjamin Barber, Strong Democracy: Participatory Politics for a New Age (Berkeley, CA, University of California Press, 2003), p. 123.

  80. 80

    Earlier versions of this paper have been presented at the University of Stirling, the University of Edinburgh, Rome LUISS University, the Australian National University and the University of Wollongong. For their very helpful comments, I thank Peter Balint, Andrea Baumeister, Emanuela Ceva, Rowan Cruft, Antony Duff, Lina Eriksson, John Horton and Ambrose Lee. Thanks also to Samuel Baker for discussing some of the ideas presented in this paper. Finally, I am very grateful to two anonymous referees from the Journal of Applied Philosophy for their valuable feedback on earlier drafts.