Your complimentary articles
You’ve read one of your four complimentary articles for this month.
You can read four articles free per month. To have complete access to the thousands of philosophy articles on this site, please
Nevin Chellappah asks whether John Stuart Mill’s famous account of free speech is still sustainable in the age of Twitter.
It is arguably the paramount value in liberal democracies and the foundation of our other liberties. The fervent endorsement of free speech by so many today can be traced back to John Stuart Mill’s reasoning in Chapter 2 of his essay On Liberty (1859). Mill made a powerful argument for allowing free speech because, he said, it is essential in the search for truth. Yet, whilst Mill had a clear conception of the end benefits of free speech, many of its modern defenders tend by contrast to see it as a prima facie good: something that should be allowed except where there is a particular reason not to. The implication is that free speech has an inherent value, not just an instrumental one, and this suggests a fundamental shift in how we understand it. Hence, in this article I will argue that the classical liberal version of free speech espoused by Mill is no longer compatible with the digital age, especially for social media.
Mill’s Argument for Free Speech
First, let me set out Mill’s account of free speech. His first concern in On Liberty is with the suppression of opinions by an authority. For him:
“The peculiar evil of silencing the expression of an opinion is that it is robbing the human race; posterity as well as the existing generation; those who dissent from the opinion, still more than those who hold it. If the opinion is right, they are deprived of the opportunity of exchanging error for truth: if wrong, they lose, what is almost as great a benefit, the clearer perception and livelier impression of truth, produced by its collision with error.” (p.19).
In response to censorship being presented as a trusted system to filter out true expressions from false ones, Mill says that there is no perfect censor. This is demonstrated by history, with past ages suppressing ideas that are now accepted to be true (the Catholic Church’s censorship of Galileo comes to mind). His emphasising that “facts and arguments, to produce any effect on the mind, must be brought before it” (p.22) illustrates Mill’s central belief in the importance of a diversity of viewpoints and the value of individual thought.
Mill’s argument for free speech could be condensed as follows:
Premise 1: The truth is valuable and people should be allowed to arrive at true beliefs
Premise 2: Freedom of speech enables people to arrive at true beliefs
Conclusion: Therefore, freedom of speech is valuable and ought to be promoted and protected
This shows that Mill narrows his focus to one specific aspect of free speech: free discussion. Free speech, according to Mill, means the freedom to express an opinion. This can be even further simplified to the freedom to assert a proposition. Therefore Mill understands speech as an action in seeking knowledge. Whilst such an appreciation of speech as a means of doing this is crucial, Mill cannot sidestep the sizeable objections that social media highlights in cases where speech performs a different action. Social media was prophesied as a digital leap in democratic speech, clearly modelled after Mill’s ‘marketplace of ideas’, expanding that into a global marketplace. But, as we now know, social media does not primarily serve for the dissemination of knowledge.
In the modern world, social media acts as the primary means for communication across borders, so we should consider which account of free speech is applicable in this landscape. It bears asking whether Mill’s definition of free speech still applies in this digital space.
Emotions versus Ideas
Social media presents three major challenges to Mill’s account of free speech. The most immediate is that it does not differentiate between an exchange of ideas and an exchange of emotional responses. This is because social media does not reward the truth value of propositions. Content is promoted based on its popularity rather than its truth. Also, the fact that social media is often used for relaxation or simple amusement means that ‘emotional’ statements – statements formed by sentiments or attitudes rather than by, say, rational argument – are often appealing. Emotional content is easier to understand and connect with, and this has led to such content dominating on digital platforms. The urge to generate emotional reactions and seek the approval of one’s followers and friends results in intellectual debate being pushed to the margins. In its place emotional claims become perceived as fact-based propositions. As Adam Moore points out (in Social Philosophy & Policy, Vol 37:2, 2021), if legitimacy is given to a social media speech act thanks to the number of likes or shares, “the gatekeeping mechanisms of quality, dedicated to the quest for truth… are irrelevant.”
Evidently, this differs a great deal from what Mill had in mind when advocating free speech. The disparity suggests that Mill’s conception of free speech is inapplicable in the digital age. But perhaps this distance derives from Mill’s idealism? For him, free speech was in order to seek knowledge which could be put in action. Mill’s defence of free speech looks rather like, as Piers Norris Turner notes, “a defense of not restricting viewpoints in frank and fair-minded public discussion” (Utilitas 33:2, 2021). Mill’s idealistic implicit belief that free speech applies primarily to academia or politics, means his understanding of it is limited.
The underlying issue from which this challenge to Mill’s account of free speech stems is the sheer size of social media’s ‘marketplace of ideas’. A global platform is far too large for productive and engaging discussion because it is affected by what Moore calls ‘content pollution’. The overabundance of content forces communication to prioritise emotional engagement over intellectual curiosity due to the preference for more ‘bingeable’ content. The Darwinian battle no longer seeks the truth, but instead attention. Mill’s argument was primarily attempting to establish the connection between free discussion and the good of humanity, but much, if not perhaps the majority, of speech on social media has little to no positive value in these terms; and yet such speech is valued. This indicates that Mill’s account of speech as an act of seeking knowledge has been powerfully damaged by social media.
However, this challenge is not insurmountable under Mill’s account. Expanding his understanding of free speech to include emotional responses that do not always lead to the truth as well as the discussion of ideas seems like a practical solution which doesn’t change the classical liberal account too dramatically. It maintains the foundation of Mill’s view whilst adding the modern dimension of valuing emotional discussion which has little or no intellectual direction. Perhaps as a child of the internet, who has always known this form of speech, I do not think that this is a fatal flaw for free speech in the digital age. Social media is formed by subjective human performances that emotionally resonate and offer glimpses of our shared humanity. I cannot help but welcome this expanded expression.
Whilst Mill may not approve, his account can survive the first challenge by recognising the two forms of discussion on social media. However, this does not resolve the problem of whether social media is a true marketplace of ideas.
Image © Venantius J Pinto 2022. To see more of his art, please visit behance.net/venantiuspinto
No True Free Market of Ideas
The second challenge posed by social media to Mill’s account of free speech is that it does not facilitate a true ‘marketplace of ideas’. Viewpoint diversity and individuality are not applauded, but are vilified and attacked.
Social media is funded through trading personal data to advertisers. Algorithms categorise opinions, and that personal data is then used, as Richard Sorabji points out, “to target users with information, or disinformation, tailored to their personality” (Social Philosophy & Policy, Vol 37:2) – in other words, to send people content which is centred on ideas they already agree with. Ideas are reaffirmed rather than challenged, and accordingly, social media groups are affirmed in their apparent objective truth. This affirmation leads to increased hostility when those groups are exposed to those who do not share their view.
This problem is only exacerbated by a lack of reasoned debate, substituted with emotional rhetoric, leading to a dilution of the meaning of words which are then hurled as terms of abuse – perhaps most prominently seen in Godwin’s Law: ‘As an argument progresses online, the chances of Nazis being mentioned increases exponentially’. A hostile and self-righteous attitude is frequently displayed to anyone with dissenting views. As a result of algorithms segmenting the audience and reinforcing opinions, areas emerge in which the voices of hate-filled moralists are tyrannical and empowered. Consequently, social media can easily collapse into a marketplace filled not with ideas, but with intellectual thuggery.
Emotional reactions are not inherently wrong, but on social media are likely to be amplified through a sort of inflammatory positive feedback. In a frenzy of indignation, ‘righteousness’ actually dehumanises digital conversation. Furthermore, the ‘debate’ is not governed by facts and arguments, and so wrong opinions do not gradually yield to fact and argument as Mill envisaged. Rather, abuse slips in online easily because social media speech already ignites emotion, and very quickly the emotion is anger.
Mill’s account of free speech is challenged as being ineffective for dealing with this. He feared social disapprobation being used as a form of censorship, with the views of the majority crowding out the opinions of the minority. Yet, this seems precisely the direction of social media, due to the lack of an adequate mechanism for facilitating the effective exchange of ideas.
However, perhaps Mill’s philosophy has other resources that can help. The famous ‘harm principle’ that he advocates elsewhere in On Liberty says that the only limits placed on personal freedom should be to prevent harm to others. His commitment to this principle has often been misunderstood because of how sparingly he is willing to use it to sanction limits to free speech. Rather than focusing on the harm that speech can produce, Mill links the harm principle to his educative and democratic aims for public discourse. He’s a strong supporter of “condemning everyone… whose mode of advocacy… [is] malignity, bigotry or intolerance” and “giving merited honour to everyone who has calmness” (p.54). This suggests that Mill could support the regulation of discussion in those terms. Hence, for Mill’s free speech to withstand this challenge, the caveat must be added that hate does not lead to true beliefs but in fact clouds judgment. This can then ensure that Mill’s argument about the value of wrong opinions remains compatible with social media, so long as opinions are civilised in their presentation and in their reaction to being challenged. Furthermore, Mill’s utilitarian ethics implies a duty to engage in constructive dialogue because it is conducive to increasing the general happiness. This appears compatible with social media and may prevent destructive, abusive conversations more effectively than his harm principle.
I believe algorithms, with their ability to categorise opinions and target advertising, can play a crucial role in this reform. They could be redirected to present a wider variety of opinions in order to form an actual idea marketplace. In doing so, we could inform peoples’ judgements without their being surrounded by abuse.
Algorithms being made more compatible with Mill’s ethical principles would centre an idea marketplace around us and so fulfill, even transcend, Mill’s vision for free speech. Thus, the classical liberal conception of free speech can survive this second challenge by reaffirming intellectual development in just this sort of way.
Anonymity Destroys Debate
The third, and most challenging, problem of social media for Mill’s brand of free speech, is the fact that it prompts anonymity.
Autonomy is often at the heart of arguments in favour of free speech, and the same arguments have been applied to social media. Freedom was considered to necessitate autonomy; but following the development of online platforms, we now regard anonymity as necessitating autonomy. This is sometimes deemed a more democratic situation, because it means that only opinions are being judged, not the person. The ability to be anonymous, at least in the sense of being detached from consequences, was a great selling point for social media. Robert Post even says, “the possibility that your digital character has more truth than your reality” suggests that social media has allowed for greater self-knowledge (Free Speech in the Digital Age, 2019).
For me these considerations recall Joel Feinberg’s four concepts of autonomy – one of them being the “the capacity to govern oneself for self-realisation within moral boundaries” (Notre Dame Law Review, 58:3, 1983). The idea that self-realisation must be an independent process falls nicely under the autonomy-building aspect of free speech and expression on social media: we are displaced from ourselves, and are instead given a new identity to discover new ideas.
For Mill, free speech enables the flourishing of ideas, which in turn leads to the flourishing of ourselves. So such arguments champion Mill’s belief that speech has good ends in terms of our self-fulfillment. However, in light of content pollution, autonomy arguments for social media are rather anemic, because quality and relevance are jeopardized by false and trivial personas. Although many kinds of expression are either autonomy-building or autonomy-protecting, it is also true that many expressions are almost entirely destructive, or, worst of all, pointless. Stanley Fish maintains that “speech in short is never a value in and of itself, but is always produced within the precincts of some assumed conception of the good” (There’s No Such Thing as Free Speech… And It’s a Good Thing Too, 1994). Here, speech is an action with purpose: it is doing something with words. The fact that much anonymised content on social media is not end-seeking also seems incompatible with Mill’s account of free speech, which as I mentioned, he basically justifies in terms of intellectual improvement.
Mill thought that protecting free speech would embolden people to offer their opinions. Furthermore, the desire to avoid social disapproval would encourage people to perfect their arguments, provide proper evidence, and promote them in a civilised manner. But shame is non-existent in anonymised conversation. This has both benefits and drawbacks. On the one hand, as Moore realises, the “connections between privacy, autonomy, and flourishing are strong and important” (ibid). On the other hand, without social consequences, people feel empowered to say just anything they want. This demonstrates a fundamental problem which Mill’s account cannot address in the digital age: ‘free speech’ does not mean ‘consequence-free speech’. Mill didn’t say much about this, because for him the real consequence was whether an opinion survived the test of the marketplace.
Speech seems inherently connected to an individual, but the ability to distance an individual from what they say to the point that it has no effect on them poses a grave challenge to free speech. So Mill’s argument cannot withstand the unaccountability problem of social media; perhaps no account of free speech can. Social media has removed the good of mental development that Mill’s free speech tries to tend towards, and accordingly, this has fundamentally changed the value of speech. Therefore, Mill’s understanding of free speech cannot survive the anonymity problem.
The effects of technology on communication are echoed in Plato’s Phaedrus. Socrates tells the legend of the Egyptian god Thoth presenting the art of writing to King Thamus as a gift to his people. Thoth says it is a remedy for the memory, but Thamus retorts that it will give only the appearance of memory – in fact it will damage people’s ability to truly learn and remember. Neil Postman urges us to take to heart this cautionary tale, but as a corrective to Thamus’s judgment, notes that “technology is both a burden and a blessing, not either-or, but this-and-that” (Technopoly: The Surrender of Culture to Technology, 1992).
It is clear that social media has influenced the nature of self-expression, from allowing emotive, non-factual discussion to take precedence over an exchange of ideas without distinguishing between the two, to empowering a self-righteous moralism deriving from algorithms that create a false sense of objectivity. Yet Mill’s classical liberal model of free speech can survive and adapt to these challenges by including emotional dialogue within its definition and emphasising calmness in expression. It is the final challenge of anonymity or consequence-free speech that poses the greatest problem to Mill, because it potentially changes the entire nature of discourse. Perhaps no account of free speech can withstand that. However, I cannot help but believe that the possible demise of classical liberal free speech is not a fatal flaw of the digital age, but instead an unintended consequence of technological development that will have to be overcome just like any other problem presented by new technology.
© Nevin Chellappah 2022
Nevin Chellappah enjoys reading philosophy, watching films – even better, trying to understand both – and boring his friends with his so-called ‘philosophical prowess’.