People have the right not to suffer disadvantage in the distribution of social goods and opportunities, including disadvantage in the liberties permitted to them by the criminal law, just on the ground that their officials or fellow-citizens think that their opinions about the right way for them to lead their own lives are ignoble or wrong. (Dworkin 1981, p. 194)
Here the reference to opportunity and liberty incorporates freedom of speech. Dworkin’s target was authorities who wished to regulate the availability of pornography because they considered it base or sinful, or thought that people’s lives would go better if they did not view pornography.
But the most philosophically elaborate formal autonomy view is provided by another philosopher, T. M. Scanlon (1972), in a quite old but much cited article. For Scanlon, respecting formal autonomy involves respecting a person’s sovereignty in deciding what to believe and what to do in light of what she hears, sees or reads; so his is an audience-based theory. In contrast to Dworkin, Scanlon is concerned with individuals who might go on to commit harms against others on the basis of speech they receive. According to him,
There are certain harms which, although they would not occur but for certain acts of expression, nonetheless cannot be taken as part of a justification for legal restrictions on these acts. These harms are: (a) harms to certain individuals which consist in their coming to have false beliefs as a result of those acts of expression; (b) harmful consequences of acts performed as a result of those acts of expression, where the connection between the acts of expression and the subsequent harmful acts consists merely in the fact that the act of expression led the agents to believe (or increased their tendency to believe) these acts to be worth performing. (Scanlon 1972, p. 213)
Suppose I calmly tell you that Jews control our country’s financial system, that Muslims are all terrorists, or that our country’s gay teachers are trying to ‘make’ our children gay. All three views are false. But this fact cannot justify censoring my speech, because it is you, as an autonomous agent, who holds the right to decide what to believe. As Scanlon puts it in speaking of someone who has heard harmful views, ‘[t]he contribution to the genesis of his action made by the act of expression is, so to speak, superseded by the agent’s own judgment’ (Scanlon 1972, p. 212). However, if you were already an anti-Semite, Islamophobe or homophobe – that is, you already had those wrongful beliefs – and I urged you to attack one of these groups, or, even worse, if you had already decided to attack them and I gave you key information helping you to do so, that would be a different matter: in such cases my speech could be legitimately restricted. But, to use one of Scanlon’s examples, Martin Luther, who nailed his 95 theses to the door of the Wittenberg Cathedral, thus starting the Protestant Reformation, could not be held responsible for the bloody religious wars that resulted from that major schism within the Christian world.
For some people, this will be counterintuitive. If you are the victim of a hate attack, you might well believe that you have the right to complain not just against your attacker but also against the person or group that fed him with prejudice and bigotry. Your attacker is morally responsible for what he did, but he acted in certain circumstances, and – even if this is difficult to prove – might have not acted in their absence. The agents responsible for engineering those circumstances, for example by creating extremist websites, arguably bear some responsibility too. Unlike Scanlon, then, we often think that speech that prompted individuals to commit harm should be regulated in some way.
A further issue with the autonomy theory is that it is relatively indiscriminating in that it does not offer especially stringent protection for certain kinds of high-value speech (see Sunstein 1993, pp. 141–3). From the perspective of autonomy, a view is a view that we have the right to receive or to express. But, like Mill, many thoughtful people today attach special value to speech about moral, religious, political, historical, scientific, cultural or artistic matters. Suppose, for example, that a government concerned about mitigating the effects of climate change banned airlines from advertising on television, radio, the Internet and print media in the same way in which, for public health reasons, it currently bans cigarette manufacturers from advertising. Compare that with a government that, concerned about its re-election prospects, sought to ban from those same media all discussion of an opposition party or movement. For everyone who values democracy, there seems to be something intuitively much more troubling about that second ban. But, from the perspective of respect for autonomy, it is less clear how the political ban is worse.
For most people who value autonomy, what really matters is not just its formal but also its substantive dimension – the ideal of individuals realising their capacity to chart their own lives. After all, it might be argued, there is little point in respecting a capacity that few people ever realise, if any at all. On the substantive view, personal autonomy is the ideal that individuals critically evaluate the cultural resources around them, including the speech of others, in order to choose and pursue those aims that they endorse. Since, in order to live self-directed lives, we also need to speak to others, the substantive view supports free speech both from the perspective of audiences and from that of speakers themselves.
There are, however, two large issues that stand in the way of a close connection between speech and substantive autonomy. One is that much speech bypasses or even subverts the capacity for critical reflection that lies at the heart of the ideal of substantive autonomy; thus, on the deontological view, it fails to respect this capacity or, on the consequentialist view, it fails to promote it. In a self-critique of his 1972 article, Scanlon draws attention to how some of the speech we are exposed to as audiences may not always help us to be ‘sovereign in deciding what to believe and in weighing competing reasons for action’ (Scanlon 1979, p. 531). ‘Expression is a bad thing’, he writes, ‘if it influences us in ways that are unrelated to relevant reasons, or in ways that bypass our ability to consider these reasons’ (p. 525). Deceptive and subliminal advertising are clear examples of speech that bypasses our rational autonomy in order to persuade us to think and act in certain ways. And so are, arguably, certain types of food labels or forms of expression that are used as nudges, to ‘[s]ystematically [exploit] non-rational factors that influence human decision-making’ (Hausman and Welch 2010). We might also think that a diet of fake news, as a result of which individuals are continually exposed to false views for which there is no evidence, undermines these individuals’ capacity to assess critically what they hear, see and read. (We return to the issue of fake news in Chapter 6.) As Susan Brison has argued, hate speech, too, can undermine audiences’ and bystanders’ capacity for autonomy by leading them to acquire false beliefs such as that they are inferior and by undermining their self-esteem, both of which damage the capacity to reflect critically on one’s surroundings (Brison 1998, pp. 326–8).
The other problem with the substantive view is that many people are not (or do not want to be) autonomous in the sense described by the ideal presented here. Consider a person who enters a closed religious order, where every aspect of her life is governed by strict rules interpreted and administered by religious elders. For such a person, free speech may still be important because it allows her to participate in religious prayer or other rituals; indeed these seem especially important and valuable forms of speech, but not because they involve the exercise of autonomous capacities. Or imagine a person who is a slave to peer pressure and craves approval from others so much that she follows the lead of her friends and family in every important life decision. Such a person does not seem to be very autonomous either, but most of us would argue that free speech is equally necessary for her. Perhaps relatively few people employ critical reflection and self-conscious choice in the way autonomy demands; nor does it seem necessarily wrong to reject autonomy. If the ideal of autonomy is ‘sectarian’ and autonomy as a fundamental good is ‘an idea about which there is much reasonable controversy’ (Cohen 1993, p. 222) in diverse societies, then according to many liberals it may not be politically legitimate for the state to protect free speech on its basis (cf. Rawls 2005 and Bonotti 2015).
Faced with this lack of connection between free speech and autonomy, a number of contemporary writers have sought to define some central interests of persons – capacities that, if realised, enable all people’s lives to flourish – which are more closely related