Piero Moraro, University of Stirling, Autonomy and Autonomy-based Duties: an argument for disobedience?

·  Autonomy and Autonomy-based Duties: an argument for disobedience?

One of the main problems in justifying civil disobedience (CD) is that CD seems to represent a way to ‘coerce’ other people’s freedom, and therefore should not be acceptable within a liberal state. The target of this paper is to analyse what ‘coercion’ means and implies. I start by comparing ‘autonomy’ and ‘freedom’, arguing that the former has priority over the latter in a liberal state. I aim to highlight that, under some circumstances, it might be justifiable to coerce someone’s ‘freedom’ for the sake of his/her ‘autonomy’. I will refer to Gerald Dworkin’s ‘hierarchical’ theory of autonomy to show that ‘coercion’ could be an incentive, rather than a threat, to one’s autonomy. Then I will discuss some aspects of Joseph Raz’s theory of freedom. I will conclude, then, by suggesting an autonomy-based justification for CD.

There is wide disagreement about the definition of the ‘autonomous agent’, and I do not intend to enter that debate here. For the sake of my argument, it suffices here to describe autonomy as ‘the condition of self-governing’: that is to say, an agent is autonomous when s/he can set and pursue her own ends[1]. This does not imply that autonomy means freedom. Gerald Dworkin offers an interesting distinction between ‘autonomy’ and ‘freedom’[2]. He takes the latter to be a local concept, and the former a global one: the scope of freedom and that of autonomy are, to this extent, different. The question of freedom is decided at a specific point in time: someone was free to do such and such at time t. At a later time, he was not free to do that. On the other hand, the question of autonomy is assessed only on the long term: it is a form of assessment that evaluates a whole way of living one’s life. Autonomy thus refers to the identification of a person with her own projects, values, goals, desires. What follows is that “freedom is only a necessary condition for autonomy. So autonomy involves more than just being free.”[3] Te idea here is that limiting one’s freedom sometimes might be the very way to promote the exercise of one’s autonomy. Dworkin’s ‘hierarchical’ theory of autonomy aims to show why, and how, this could be the case[4].

Dworkin: ‘Second-Order’ Identification

I pointed out above that autonomous individuals are able to reflect on their own decisions, desires, habits, and to form preferences about them. These involve also reflection on what Dworkin calls ‘second-order desires’. For example, someone who has a desire to smoke may also have the desire that s/he have not that desire. S/he can also desire that her/his motivation to act be different:

To give another example, a person might desire to learn to ski. He might believe there is no further motivation or he might believe that what causes the desire is the wish to test his courage in a mildly dangerous sport. Suppose he is now led to see (correctly) that he desires to ski because he is envious of his brother who has always excelled in sports. Having recognized the source of his desire he can now either wish he were not motivated in this way or reaffirm the desire. If the latter, then he is acting authentically in that he identifies himself as the kind of person who wants to be motivated by his envy.[5]

This is what Dworkin calls ‘second-order identification’ with one’s own desires. If autonomy implies such self-identification, i.e. seeing one’s plans and projects as one’s own, it requires that the agent identify not only with her/his desires, but also with the motivations behind those desires. It is only when one identifies also with those motivations, with the influences behind one’s own desire, and sees oneself as the kind of person who wishes to be moved in that particular way, that the motivations are to be identified as “one’s own”. It follows, then, that if one resents being motivated in certain ways, i.e. would like to be a person motivated by different influences, then the motivation, though causally effective, would not be viewed as “his” or “her” own[6]. This is what Dworkin calls authenticity, that is, the identification of one’s second-order desires as one’s own, and this he takes to be the characteristic of an autonomous agent. There is, furthermore, a second condition, what he defines as procedural independence. This refers to the fact that the second-order identification a person makes might have been itself influenced by other persons, in such a way that we do not view it as being his or her own: the way in which the motivation is formulated matters as well. Thus, Dworkin’s formula for autonomy is as follows:

A person is autonomous if he identifies with his desires, goals and values, and such identification is not itself influenced in ways that make the process of identification in some way alien to the individual[7].

The role of procedural independence in the autonomous subject has an important consequence: it follows that not every way of influencing people constitutes an interference with their autonomy. Some of these influences amount to threats, and represent, strictly speaking, the kind of “influences” on one’s desires that, for example, Isaiah Berlin was wary of[8]; others, on the other hand, may constitute incentives for that person’s autonomy. What Dworkin argues is that, while some ways of influencing people’s reflective and critical faculties may subvert them, others may improve them. Coercing individual “local” freedom might promote his/her “global” autonomy, i.e. his/her condition as a self-governing individual.

How do we reconcile this with the idea of ‘respecting’ someone’s autonomy? According to Kant, for example, we ought to treat individuals never merely as means but also always as ends in themselves: this respect follows from the autonomous individual’s being capable of creating ends that can be universalized. Does that imply we should protect someone’s autonomy from any form of violation? It is not so straightforward that “respect” entails “never violating” someone’s autonomy. If we grant that autonomy should be valuable ‘for its own sake’[9], then one may ask if it constitutes a moral constraint on our behaviour, or rather a goal, the promotion of which we have to aim for. According to the former view, we should never infringe upon someone’s autonomy: seeing someone as capable of being autonomous, we should never limit that capacity. However, and this seems to be suggested by Dworkin’s conception of autonomy, being in the condition of self-government requires a higher-level identification of the agent with his/her own desires: and this might justify some form of “incentives”. The idea is that, rather than taking the individual’s autonomy as being a constraint on action ipso facto, we should aim to promote the goal of his or her autonomy in the long term.

Therefore, “respect” might be displayed through aiming to promote someone’s autonomy, in recognition of the fact that she might not be fully in the condition to self-legislate. This seems to be the way to interpret Dworkin’s claim that some forms of ‘influences’ upon people’s desires (i.e. upon their own ends) do not constitute threats: rather they are incentives to people’s condition of self-government. When such influences are motivated, at least in part, by the desire to promote ends that they might have at that time (while being unable to see them), individuals are not treated as mere means to an end. It follows that respect may imply taking autonomy as a value to be maximized over time.

So far, two aspects have been highlighted. First, there is more to autonomy than just ‘freedom from coercion’: in fact, being in the condition of self-government might require more than simply ‘being free from’. In fact, it seems that some forms of influence might not constitute threats, but rather incentives, to someone’s autonomy. Second, ‘respect’ for autonomy might allow for ‘influencing’ someone’s autonomy when that constitutes an ‘incentive’ to his/her autonomy. These two issues play an important role in Joseph Raz’s autonomy-based theory of freedom, which contributes to the thought that autonomy does not confine itself to ‘being free from coercion’. I now turn to discussing the main points of Raz’s concept of autonomy.[10]

Raz: The Morality of Freedom

Raz identifies three ‘conditions of autonomy’ that must be fulfilled for the agent to be able to lead an autonomous life: (a) appropriate mental abilities, (b) adequate range of options, and (c) independence[11]. By (a) Raz means that the autonomous agent must be in possession of a minimum of rationality, i.e. the mental faculties to set one’s own goals, the ability to comprehend the means required to realize them, etc. More interesting in Raz’s account is (b): the idea that ‘autonomy’ requires the availability of an adequate range of options to choose from[12]. In order to understand this point, we should have a glimpse at Raz’s notion of well-being[13]: a person’s well-being depends on her being the maker or the author of her own life, and on the availability to her of a multiplicity of valuable options[14]. There is a direct connection, therefore, between having a valuable life and being autonomous. A life in which an agent is not in the condition to choose would be much less valuable than one in which s/he is. To be autonomous a person must not only be given a choice, but s/he must be given an adequate range of choices. Someone whose decisions are extracted from him/her through coercion is not acting autonomously: equally, someone who is paralysed and cannot take advantage of the options available to him/her lacks autonomy. If the autonomous life is about choosing, then, guaranteeing someone’s autonomy entails providing the individual with an adequate range of options from which to choose. By ‘adequate’ Raz emphasizes not the ‘number’, but rather the ‘variety’: “[a] choice between hundreds of identical and identically situated houses is no choice, compared with a choice between a town flat and a suburban house”[15]. Furthermore, ‘variety’ in the strict sense is not enough either: choosing between a variety of morally repugnant actions does not qualify as ‘autonomous choice’. If I am faced with a choice between applying for a PhD at the university, or killing someone, then I am not autonomously choosing what to do: for Raz, the choice between good and evil is no choice at all.

What follows from this, once again, is that in order to qualify as an autonomous agent, one has to focus on much more than simply freedom from coercion[16]: on the one hand, autonomy implies the availability of a range of options from which to choose; on the other, these options have to be morally good ones. This sheds light on Raz’s perfectionist conception of the State: some forms of life are, for Raz, objectively better than others, and the State is justified in promoting those that contribute to individual autonomy, and discourage those that conflict with it. Even more, according to Raz there are specific duties that follow from this conception of autonomy

Since autonomy is valuable as a constitutive element of one’s own well-being, there is reason for everyone to make oneself and everyone else autonomous:

[b]ut it is the special character of autonomy that one cannot make another person autonomous. One can bring the horse to the water, but cannot make it drink. One is autonomous if one determines the course of one’s life by oneself. This is not to say that others cannot help, but their help is by and large confined to securing the background conditions which enable a person to be autonomous.[17]

According to Raz, it is not the case that the State simply should promote autonomy: he argues thye State has duties to do so. He points to three of them, corresponding to the three conditions of autonomy already mentioned. One is the duty to guarantee ‘independence’ of choice: the State has an autonomy-based duty to refrain from coercing or manipulating someone into choosing a particular way of life. There is then a duty to help creating the inner capacities required for the conduct of an autonomous life. Some of these- writes Raz- are cognitive capacities, like reasoning abilities, the power to absorb and use information, and so on. Others concern one’s emotional and imaginative make-up. Others still regard physical abilities, and character traits essential or helpful for an autonomous life.

There is a third autonomy-based duty, which acquires particular relevance in the present discussion of autonomy in a liberal society: this is the duty to provide others with an adequate range of options from which to choose. The State, according to Raz, has the duty to provide individuals with a range of morally good alternatives. It has a duty to create, for its citizens, the conditions for making choices, which as we noticed already is, for Raz, constitutive of a life worth living[18].

Once again, it appears that there is more to do to secure the conditions of autonomy than simply refraining from coercing others, i.e. guaranteeing negative freedom. There is a principle of autonomy, as Raz calls it, which requires the State to go far beyond the negative duties of non-interference. According to this view, the value of “negative” freedom is conditional on its contributing to autonomy: it may affect someone’s autonomy, but it need not do so. Some options one is better off without having. So negative freedom is valuable inasmuch as it serves “positive” freedom: the latter is more directly connected to autonomy as the capacity to choose one’s own goals.

From this, Raz argues that the State might be justified in coercing people, when coercion aims at fostering their autonomy. Raz’s argument is here not entirely clear[19]: however, it seems to hinge mainly on the intuition that the State has a duty to discourage, or even forbid, morally wrong options. Thus, Raz argues that coercion is not necessarily bad, since it might be a way to promote the well-being of the individual. This need not result in a tyrannical regime that tells people what to do, for the liberal state cannot be monist, promoting one way of life over all the others: quite the contrary, the State ought to allow for value-pluralism, which is the necessary condition for choice.