Tom Bradshaw

John Stuart Mill: Freedom of expression and harm in the 'post-truth' era

This paper contends that John Stuart Mill's arguments for freedom of expression – despite being first published in 1859 - remain a powerful framework through which contemporary issues of free speech can be explored and taught. As part of an analysis of Mill's On liberty, the notion of intolerance being the default condition of mankind and restrictions on free speech being a trans-generational wrong are delineated. The issue of on-campus censorship is discussed, as are the prominence and causes of self-censorship in both the media and wider society. The usefulness of Mill's On liberty as a means of exploring these contemporary problems of free speech is expounded, particularly the enduring usefulness of his Harm Principle, and a phenomenon that is called the Paradox of Liberal Inheritance is identified. The concept of alethic disruption is developed to investigate the emergence of 'post-truth' news, and a Mill-inspired response to this is outlined. It is argued that Mill articulates principles that modern media students can use as an accessible means of approaching questions of media ethics, and that On liberty also makes powerful points about the etiquette of intellectual debate.

Keywords: John Stuart Mill, freedom of expression, harm, 'post-truth' era

Introduction: Freedom of expression - old questions, new questions

What are the limits of freedom of expression in a democracy? Is there even a limit? If there is one, who polices it - the legal system, public opinion or a hybrid of the two? Are either consequentialism or deontology adequate ethical frameworks through which to produce coherent answers to issues about free expression? Since Milton's 17th century plea for unlicensed publication, Areopagitica, the first three of these questions have been defining ones for the West (Berlin 2012, Cohen-Almagor 2005, O'Rourke 2003). And the importance of all four questions has been underscored by incidents in both newsrooms and the corridors of power during the opening years of the new millennium. The death threats and killings that followed the publication of cartoons of the Prophet Mohammed in the Danish newspaper, Jyllands-Posten (2005-2006) and the French satirical magazine, Charlie Hebdo (2015), illustrate that freedom of expression is not just a matter for the debating chamber, but is potentially a matter of life and death to members of the media and wider society (Rose 2006).

Elsewhere, the prospect of state-backed press regulation in the UK arising from the Leveson Inquiry (2011-2012) prompted vociferous opposition from the mainstream media, who have argued that such a move is tantamount to pulling away a pillar of open society that has been steadily buttressed since Milton's 17th century essay (Nelson 2016). Yet freedom of speech is far from being an absolute right in the UK and is limited by laws of defamation and contempt and, until as recently as 2008, blasphemy. Moreover, as Frost (2016) explains, a range of legislation has a restrictive effect on what can be broadcast or published in the UK on grounds of offence, from the Obscene Publications Acts 1959 and 1964 to the Race Relations Act 1976 and the Communications Act 2003. The free speech ramifications of such legislation are contentious - both for the media and wider society - as illustrated by the variety of mainstream polemics about freedom of expression (Garton Ash 2016, Hume 2015).

Long-established questions about the boundaries of free speech, therefore, remain prevalent. But to these questions can be added another layer of complexity. This complexity is provided by what can be termed the alethic disruption within the emerging media ecosystems of the 21st century that are dominated by social media and the algorithms that power the content distribution on those social media platforms (Knight 2013, Smith 2016). It has been contended by the new editor of the Guardian that in a 'post-truth' culture propelled by social media, the currency of online information is no longer truth but virality (Viner 2016). Audience engagement (by which is meant high unique visitor numbers and the sharing of content) becomes the altar on which 'good' journalism is potentially sacrificed. On this understanding, what engages people online and prompts them to consume and share content is not its veracity but its 'affective' - or emotional - power (Hermida 2016). This leads to the propagation - unwittingly, but sometimes wittingly - of misinformation in an era in which 'facts become secondary to feeling; expertise and vision to ersatz emotional connection' (Smith 2016). Truth is in some instances relegated to being an optional extra. The prevalence and traction of 'fake news' in the coverage of the 2016 US Presidential election and the Brexit referendum is cited as an example of truth's relegation (Hermida 2016, Viner 2016). Reflecting on the US Presidential election, Hermida describes how Donald Trump's supporters 'became the media themselves, spreading and amplifying subjective and emotional affective news - news designed to provoke passion, not inform'. This is 'affective news', a diet of content that is 'designed to stir up passions, feed prejudices and polarise publics'. Moreover,

People will share false information if it fits their view of the world. Even if some don't quite believe it, they will share an article with the aim of entertaining, exciting or enraging friends and acquaintances. Fake news spreads so fast that potentially hundreds of thousands of people could have seen it by the time it gets debunked (Hermida 2016).

The growth of 'fake news' and how to stem its flow - or at least nullify its penetration - has become a central concern for world leaders as well as social media company executives, with the integrity of democracy itself potentially being in jeopardy (Yuhas 2016).

The concept of alethic disruption is not equivalent to the concept of 'post-truth', but the post-truth milieu should instead be regarded as the most recent and vivid illustration of alethic disruption. Alethic disruption is the wider phenomenon of the networked society's increasingly strained and complex relationship with social reality and how the truth of that reality is to be mediated, grasped and verified in a multimedia world. Churnalism (Davies 2008) is another manifestation of alethic disruption. Churnalism - the passive, desk-bound regurgitation of content by journalists from press releases or rival media - has particularly affected newsrooms and social media channels and, as a consequence, those platforms' audiences. Churnalism and the emergence of post-truth fake news illustrate the evolution of alethic disruption.

What implication does the most recent aspect of alethic disruption, the post-truth milieu, have for freedom of expression? Many of the traditional arguments for unfettered - or at least largely unfettered - free expression have been developed on the basis that the free flow of debate enables truth to emerge (Milton 1973 [1644], Mill 1989). This was the thought motivating one of the most memorable figures of speech in Areopagitica:

And though all the winds of doctrine were let loose to play upon the earth, so Truth be in the field, we do injuriously by licensing and prohibiting to misdoubt her strength. Let her and falsehood grapple; who ever knew Truth put to the worse, in a free and open encounter (ibid: 35).

But if the evolution of alethic disruption means that the emergence of truth is no longer of primary importance to a proportion of online audiences then where does that leave those old arguments upon which free speech has been founded? Is such a Miltonic argument just a quaint relic from the era of the quill that is rendered redundant by the age of the meme?

There is a further element that complicates the contemporary debate around freedom of expression. This is the issue of censorship and self-censorship in the very places where freedom of thought is - or is often assumed to be - sine qua non: universities. In November 2016, the students' union at City University London - a university with one of the most distinguished schools of journalism in the UK - voted in favour of a campus ban on the Sun, Daily Mail and Daily Express. The motion, 'Opposing fascism and social divisiveness in the UK media', stated that the ban could be extended to other media outlets and that the three specified titles were 'merely used as high-profile examples' (Sweney and Jackson 2016). In passing the motion, City was the latest UK students' union to approve a campus-ban on national press titles.

Online magazine Spiked has ranked universities on the basis of their restrictions on free speech, awarding each university a 'red', 'amber' or 'green' status depending on the policies and actions of both students' unions and university administrators (Slater 2016). The 2016 survey had a sample of 115 universities. While a green rating is conferred on institutions which place no restrictions on lawful free speech, a red rating is for those universities which are 'hostile' to freedom of expression by mandating 'explicit restrictions on speech, including, but not limited to, bans on specific ideologies, political affiliations, beliefs, books, speakers or words' (ibid.). The amber rating is for an institution which 'chills free speech and free expression by issuing guidance with regard to appropriate speech and conduct' (ibid.). The 2016 Free Speech University Rankings (FSUR) refers to an 'epidemic of campus censorship'. Ninety per cent of institutions were given either a red (55 per cent) or amber (35 per cent) grading for 2016, an increase of 10 per cent on 2015. Thirty students' unions have banned newspapers, 25 have banned songs, 20 have banned societies, and 19 have banned speakers or events. The coordinator of Spike's rankings, Tom Slater, argues that such policies are inimical to universities' quest for discovering the truth. 'Today, in a time when campus bureaucrats see students as too vulnerable - or too easily led - to listen to difficult ideas, the entire purpose of the academy is being undermined, and the bar for censorship is only getting lower' (Slater quoted in Ali 2016).

While Spiked's language is arguably unhelpfully sensationalist - references to an 'epidemic' and 'campus bureaucrats' is redolent of tabloidisation - it is the case that a new lexicon now frames the debate over freedom of expression in universities. 'Safe space', 'trigger warning' and 'no platforming' are terms frequently used by universities to justify restrictions on the expression of controversial views or content. And they have led to claims that current students, as part of 'Generation Snowflake' (Fox 2016), are being cosseted and deprived of robust intellectual development (Bromwich 2016, Scruton 2016). The debate around this issue has also reached mainstream literary culture. A notable example is a novel by a Booker Prize-winning author which adopts the conceit of having as a narrator an unborn child with a satirical tongue. On the education he can expect, the foetus-narrator writes:

I'll be an activist of the emotions, a loud, campaigning spirit fighting with tears and sighs to shape institutions around my vulnerable self. My identity will by my precious, my only true possession, my access to the only truth. The world must love, nourish and protect it as I do. If my college does not bless me, validate me and give me what I clearly need, I'll press my face into the vice chancellor's lapels and weep. Then demand his resignation (McEwan 2016: 146).

Social commentators have echoed the perspective of McEwan's foetus, criticising the safe spaces and trigger warnings of the modern university campus as a form of intellectual sabotage. A vigorous case is presented by Scruton, who argues that open, Western society is itself being undermined by what he perceives to be on-campus censorship.

Those who wish to maintain the student mind in a condition of coddled vulnerability, unhardened by opposition and unpractised in argument, now police the campus, with the result that these places which should have been the last bastion of reason in a muddled world, are instead the places where all the muddles come home for nourishment. The example vividly illustrates the way in which the attacks on free speech can go so far as to close off the route to knowledge. And in the end that is why we should value this freedom, and why John Stuart Mill was so right to defend it as fundamental to a free society. Without it we will never really know what we think (Scruton 2016).

Scruton's reference to J. S. Mill at the end of the passage suggests the enduring power of Mill to inform debates around freedom of expression. This paper is focused on extending this idea by exploring in detail the manifold ways in which Mill's ideas, particularly those from On liberty, can be used as a trenchant means of analysing contemporary issues of free speech, even though On liberty was first published in 1859. As argued above, an up-to-date discussion of the boundaries of freedom of expression needs to acknowledge and incorporate both the issues posed by alethic disruption and contemporary manifestations of censorship in Western society, one species of which is the climate of on-campus restriction. This paper uses Mill as a prism through which to explore the on-going debates about offence and harm, and proposes what is called the Paradox of Liberal Inheritance. This paradox is focused on exploring how many of the beneficiaries of Mill's powerful vision of a liberal, tolerant society in which people are free to pursue what he termed 'new and original experiments in living' (1989: 81) are now attempting to stifle the very freedom of expression that is a precondition of such a tolerant society. In addition, the paper will explain why the study of Mill's ideas is vital for modern students of journalism and media ethics. It is argued that Mill is a powerful lens through which tomorrow's custodians of the flow of information can consider the limits of freedom of expression and media regulation.

The paper also proposes a potential solution to the issues of misinformation arising from alethic disruption, arguing with Jarvis (2016a, 2016b) that the appropriate response by both the media and regulators is not to attempt to shut down the various channels of (mis)information but rather to target those same channels with truth. This, it is argued, is a Millian response that is firmly entrenched in the tradition of English liberal thought. To rephrase it in a Miltonic idiom, such a response to the rising tide of fake news and post-truth content is to enable 'Truth to be in the field' so it at least has a fighting chance to put falsehood 'to the worse, in a free and open encounter'.

On On liberty

As a founding credo of modern liberal thought, On liberty has been - and continues to be - the focus of debate and re-interpretation (Berlin 2012, Cohen-Almagor 2005, Gray 2012, O'Rourke 2003, Rees 2012, Ryan 2012, Steel 2012). While this paper is not the place to contrast and evaluate all these interpretations, it is worth expanding on two points, neither of which has arguably received enough attention in the secondary literature. The first is that Mill argues that intolerance - and by implication censored speech - is the default condition of human societies. The second is that restricting free speech is what can be termed a trans-generational wrong, affecting not just the present generation but posterity too.

On liberty is an ardent plea for the individual to be free from constraints imposed by both the state and by majority opinion, where those constraints go beyond restricting behaviour that causes harm to other people. Freedom of expression is one facet of this wider freedom that Mill regards as so vital for people to flourish, both individually and collectively. But Mill does not contend that humankind has a benign nature, if only it were allowed to be free. Indeed, on occasion Mill's statements regarding human nature possess an almost Hobbesian tone, as when he claims that 'the disposition of mankind, whether as rulers or as fellow-citizens, to impose their own opinions and inclinations as a rule of conduct on others, is so energetically supported by some of the best and by some of the worst feelings incident to human nature, that it is hardly ever kept under restraint by anything but want of power' (1989: 17).

Tolerance, moreover, does not come easy to humans, 'so natural to mankind is intolerance in whatever they really care about' (ibid: 11). The emergence of toleration is for Mill, therefore, a central element to the emergence of civilisation. As such, a society which acknowledges the legitimacy of his Harm Principle - that 'the only purpose for which power can be rightfully exercised over any member of a civilised community, against his will, is to prevent harm to others' (Mill 1989: 13) - is one where the thirst for power is overcome, or at least contained. Freedom of expression is thus an unnatural position; acknowledging the right for another person to say something that is repellent or hateful to me, or which I believe to be plain wrong, is something that goes against the grain of human nature. And for that reason, once achieved, the acknowledgment is all the more precious. Another way of phrasing it is to say, as Bromwich does, that free speech is an 'aberration' and that 'In most societies throughout history and in all societies some of the time, censorship has been the means by which a ruling group or a visible majority cleanses the channels of communication to ensure that certain conventional practices will go on operating undisturbed' (Bromwich 2016). For Mill, convention is a deadening, creativity-sapping burden - referred to him in one place as the 'despotism of custom' (Mill 1989: 70) - and freedom of expression is a means of weakening its influence.

A restriction on freedom of speech does not simply create an absence of opinion; it is an act of intellectual deprivation. Mill argues that it is a restriction on the spread of intellectual capital that can deprive subsequent generations. As such, it can be termed a trans-generational wrong. In an attempt to convey his point with as much rhetorical force as possible, Mill invokes imagery of physical violence, describing censorship as a robbery afflicting not just the present generation but also future ones.

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 those who hold it. If the opinion is right, they are deprived of the opportunity for 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 (Mill 1989: 20).

Mill's commitment to free speech stems from a broader commitment to diversity - diversity of belief, diversity of lifestyle, diversity of thought - which is connected to his belief that diversity enables individuals to flourish, which in turn enables individuals to achieve a state of happiness (or as Mill, reflecting his utilitarian background, often refers to it, utility). As commentators have noted, an underlying theme of On liberty is how diversity is essential to progress (Collini 1989: xiv). This animating sprit of the essay - and of Mill's thinking in general - is powerfully captured by Berlin:

What he [Mill] came to value most was … diversity, versatility, fullness of life - the unaccountable leap of individual genius, the spontaneity and uniqueness of a man, a group, a civilization. What he hated and feared was narrowness, uniformity, the crippling effect of persecution, the crushing of individuals by the weight of authority or of customs or of public opinion (2012: 134).

It is the dissident and heretic, therefore, who need protecting. For Mill, society has a duty to enable those who hold minority opinions to feel they can express them without fear of being silenced.

Self-censorship and the 'spiral of silence'

Before exploring in detail how Mill's ideas in On liberty can illuminate contemporary issues of freedom of expression, it is important to acknowledge another important aspect of contemporary communication in the West, which is the extent and nature of self-censorship in the social media-driven era. There is a significant corpus of evidence to suggest that self-censorship is prevalent among both the news media (Preston 2009, Rose 2006, Sefiha 2010, Sturges 2008) and the public more generally (Dans 2014, Das and Kramer 2013, Hampton et al 2014). This is, perhaps, in some ways counter-intuitive, because self-censorship would seem to be an activity at odds with the click-happy culture of 'over-sharing' that some believe is characteristic of digital-native social media users (Bromwich 2016). However, self-censorship and over-sharing are arguably two sides of the same coin, both being techniques relating to the development by social media users of an online persona.

But why be so concerned by self-censorship, a slippery phenomenon which, by definition, is difficult to quantify because its defining characteristic is an absence - namely, silence? Self-censorship is important from a Millian perspective because, for Mill, the muzzling of one's own opinion - whether through fear of upsetting conventional or majority opinion, or through fear of provoking a response from the state - is an evil, in so far as it inhibits the spread of ideas and inhibits diverse opinion (and thereby diversity itself). For Mill, freedom of thought and freedom of speech/expression are intellectually indivisible; inhibit one and you inhibit both (1989: 17). As such, self-censorship - and the reasons that cause it to happen - are another means by which the intellectual life of present and future society is impoverished. Sturges (2008) offers a quadripartite taxonomy of the reasons for self-censorship, the first of which - the fear of breaking the 'constraints of conformity' - is phrased in distinctly Millian terms, invoking as it does the spectre of the tyranny of majority opinion.

Self-censorship is defined by Sedler as 'the decision by an individual or group to refrain from speaking and the decision by a media organization to refrain from publishing information' (2012). While a useful starting point, this is too narrow a definition. It is not just 'speaking' that individuals refrain from, but publishing social media updates, while media organisations might refrain from broadcasting as well as 'publishing'. Writing from a US perspective, Sedler regards self-censorship as compromising the values of the First Amendment due to it being a phenomenon that inhibits the dissemination of information or ideas to the public.

However, one response to this is to argue that most instances of self-censorship in the social media age are something less concerning - they are merely a form of self-defence that prevent one's remarks being misconstrued. This is in the position of Dans who, reflecting on his own use of Twitter, says that his behaviour on the platform was very different when he had 200 followers to when he had tens of thousands - and that the explanation for the difference is 'simply one of survival' (2014). He contends that self-censorship is inherent in the way social media functions.

Anybody who has spent time sharing information on a social network understands the dynamic, and that self-censorship is alive and well: as one's perceived or real audience grows, the amount of information about ourselves that we are prepared to share diminishes. … The fewer people I am talking to, and the better I know them, the less I have to explain myself in detail, just to make sure there is no chance of any misunderstanding (ibid).

The autoethnographic, qualitative nature of Dans's reflections on the nature of self-censorship on social media contrasts with the quantitative data gathered by Das and Kramer (2013). Through data provided by Facebook, and gathered from 3.9 million users over a 17-day period, their study attempts to capture the extent to which users of Facebook self-censor at the 'last minute' (2013: 120). Their study defines self-censorship in this social media environment as the failure to post a status update that has been drafted. However, this is a questionable definition, as a user might opt not to post an update for a host of reasons other than self-censorship - such as lack of time, the realisation of multiple mistakes of grammar or spelling etc. However, in a similar manner to Dans's notion of self-censorship being a means of digital 'survival', Das and Kramer introduce the concept of how self-censorship on social media can preserve a user's 'social capital' by not alienating their online friends, or at least a subset of them.

As Hampton et al discuss, the hope of some social media pioneers was that the proliferation of digital platforms would open the channels of communication so that those holding minority views would be willing to voice their opinions, in a way that they had not been in the pre-internet era due to the 'spiral of silence' phenomenon (2014). The spiral of silence tendency refers to people's reluctance to publicly air their views on policy issues when they think their view is a minority opinion. However, research by the Pew Center in the United States has provided evidence that the internet has achieved no such thing. The research, which consisted of a survey of more than 1,800 adults, concluded that the growth of social media platforms had not reduced the spiral of silence tendency and that people are less willing to discuss policy issues on social media than they are in person. Moreover, in both face-to-face and digital settings, people are more prepared to share their views if they believe their audience will be in agreement (ibid.).

It is important to make the distinction between censorship of facts and censorship of opinions. Research into the extent of self-censorship among the general public can implicitly focus on self-censorship of opinion (Das and Kramer 2013; Hampton et al 2014), while research into self-censorship by the media often - again implicitly - focuses on self-censorship of facts (Preston 2009, Sefiha 2010), although not always (Rose 2006, Sturges 2008). But what constitutes the greater wrong - censorship or self-censorship? Sturges is unequivocal. He contends that self-censorship is 'much more insidious' because 'if others suppress our freedom of expression it is bad, but if we allow ourselves to censor our own opinions it is worse' (2008: 256). Such a position is implicit in the argument of Flemming Rose, culture editor of Danish newspaper Jyllands-Posten, who oversaw the publication of the cartoons of the Prophet Mohammed that resulted in death threats. Explaining his reasoning behind the publication of the contentious cartoons he wrote:

I commissioned the cartoons in response to several incidents of self-censorship in Europe caused by widening fears and feelings of intimidation in dealing with issues related to Islam. ... Our goal was simply to push back self-imposed limits on expression that seemed to be closing in tighter (Rose 2006).

The notion of resisting 'self-imposed' limits in order to express a view is in the Millian tradition. Biting one's tongue for the sake of 'survival' (Dans 2014) or in order to foster 'social capital' (Das and Kramer 2013) would be anathema to Mill. While arguments in On liberty suggest that he would probably regard a phenomenon such as the spiral of silence as a naturally occurring one given humankind's innate intolerance (Mill 1989: 17), he would not regard it as inevitable but, instead, a symptom of the fact that the roots of tolerance in society were not extending deeply enough. Contemporary manifestations of censorship and self-censorship, and the appropriate response to them, are thus potentially amenable to a Mill-inspired analysis. This paper now turns to the most powerful concept that Mill offers those who seek to understand current issues of free expression.

The Harm Principle revisited

Mill's position on freedom of expression flows from what has become known as his Harm Principle, a deceptively simple ethical proposition which has generated much debate regarding its practical application. The principle states that 'The only purpose for which power can be rightfully exercised over any member of a civilised community, against his will, is to prevent harm to others' (Mill 1989: 13). Mill makes this statement because of an absolute commitment to self-determination, in so far as that autonomy does not harmfully affect others. 'Over himself, over his own body and mind, the individual is sovereign' (ibid).

Since the essay was first published, critics have claimed that it is bogus to suggest there is a workable distinction between actions that affect only the individual agent and actions that affect others. The history of this line of argument is well-documented by Rees (2012), who quotes a leading article from The Times Literary Supplement of 10 July 1948:

The greater part of English history since his [Mill's] day has been a practical commentary on the fallacy of this distinction. No action, however intimate, is free from social consequence. No human being can say that what he is, still less what he does, affects no one but himself (Rees 2012: 171-172).

Whether or not the distinction between purely private actions or other-regarding actions can be soundly drawn is a question that goes to the heart of debates about liberal versus conservative approaches to social policy. But the distinction does have a plausibility to it, a plausibility which Mill helps establish by differentiating between 'definite' damage and 'contingent' damage. No man should be punished for simply being drunk, he argues, but a policeman, for example, should face legal and social sanction for being drunk on duty because there is 'a definite damage, or a definite risk of damage, either to an individual or to the public' (1989: 82). But where there is 'contingent' injury, by which Mill means behaviour that does not breach a specific public duty or cause hurt to anyone but himself, 'the inconvenience is one which society can afford to bear, for the sake of the greater good of human freedom' (ibid).

'Damage' here is being used as a synonym for 'harm', and in providing an analysis of contemporary issues relating to freedom of expression it is vital there is lexical precision. The key distinction that needs to be drawn is between 'harm' (or 'damage') and 'offence', and it has been argued that another important distinction is that between freedom of speech and freedom of the press. Cammaerts (2015) contends that the two freedoms are not equivalent, with freedom of the press arguably carrying a greater burden of social responsibility than the former. He further argues that freedom of expression 'is not necessarily a primary right in all circumstances, it has to be balanced out with other rights and protections, for example the right not to be discriminated against, the right not to be racially abused' and uses the concept 'Intricate Freedom' to capture the complexity surrounding free expression (ibid). It is a contention of this paper that, while freedom of expression is indeed intricate in this way, a Mill-inspired delineation of the distinctions between offence and harm continues to provide a useful frame through which to disentangle debates around freedom of expression.

Frost describes the issue of harm and offence as one of 'constant controversy', contending that freedom of expression 'must include the right to offend, but does it include the right to harm, and when does offence turn to harm?' (Frost 2016: 198). Underpinning Frost's question is the issue of how exactly we define 'harm'. But despite the clear need for precision in drawing distinctions between harm and offence, the line between the two concepts still often appears blurred. For example, in a recent study of how members of the public understand offence (Das and Graefer 2016) the authors write:

On the one hand are those who champion free speech, who argue that we all have the right (some say, the duty) to express sometimes unpopular opinions without fear of censure. On the other are those who say that with that right comes a responsibility not to needlessly offend. For many who take this latter viewpoint, mocking religious symbols of two religions equally may seem equivalent on paper. But if one of those religions links to a culture, and very often an ethnicity, that already faces widespread discrimination and hardship, then perhaps free speech has crossed the line into offensiveness (ibid).

The use of the word 'offensiveness' here is confused with 'harm' or 'harmfulness'. Free speech does not 'cross a line' into offensiveness; offensiveness is part of the sphere in which free speech is deployed - the ability to say things that offend is what makes speech 'free'. The line that should be avoided being crossed is the line into harmfulness. Das and Graefer later acknowledge this when they write that 'the umbrella of “harmful and offensive material” needs more nuanced, focused and critical research' (2016). But where does the distinction between the two lie, and what sort of harm is it that justifies restriction on free expression? Mill adopts a hard line on this point.

Steel believes Mill sets a high threshold for the circumstances in which the state can legitimately restrict freedom of speech. This is because, on his reading, harm is understood by Mill to mean a physical hurt; perceived emotional or mental 'harm' are not sufficient to justify restriction.

For Mill, these circumstances are those in which life and liberty come under an imminent threat of harm. … Harm here … is understood as physical assault or at least the threat of assault on a person or property. As such for Mill, the limits of freedom of speech are very specific indeed (Steel 2012: 22).

Collini, too, sets the threshold high, summarising Mill's position as being one where the only legitimate ground for restricting a person's activities is that they 'are likely to produce definite harm to some identifiable other person or persons' (Collini 1989, xiv, my emphasis). The adjectives 'definite' and 'identifiable' are important here. On Collini's reading of the Harm Principle, vague notions of possible harm being caused to some potential group or other are not acceptable to justify state interference in individual behaviour. The harm needs to be clear and specific. Mill's principle might not always allow immediate, definite conclusions to be drawn about the rightness or wrongness of societal interference over individual behaviour. But Mill's point is that the onus of proof for the occurrence of harm lies with the institution or person proposing the restriction (ibid: xvii).

Mill's arguments here relating to harm and freedom of expression are not conclusive for the simple reason that what should be meant by 'harm' can be - and is - contested, for all the attempts at lexical precision. Cohen-Almagor believes Mill deals with the limits of free speech in a 'hasty' manner, claiming that, like other philosophers in the liberal tradition, including Milton, Dewey and Rawls, he wants to focus on 'principles, not the exceptions to them' (2005: 15-16), while Berlin alleges that 'rigour in argument is not among his [Mill's] accomplishments' (2012: 145). However, even Mill's critics, such as Cohen-Almagor, who argue that hurt should in some cases encompass emotional impact (or 'offense to sensibilities'), frame their position in Millian terms, as when Cohen-Almagor states that 'generally speaking - there is a need to strike a balance between the right to freedom of expression and harms that result from a certain speech. It is argued that the right to exercise free expression does not include the right to do unjustifiable harm to others' (2005: 6). Despite 'harm' and 'offence' remaining contested terms, it remains the case that Mill frames the debate with clarity.

'Experiments in living' and the Paradox of Liberal Inheritance

Through On liberty's espousal of diversity, toleration and support of dissenting voices, Mill produced arguably the most powerful credo of modern liberal thinking (Berlin 2012; Collini 1989). This credo helped set the intellectual and social climate that has since enabled previously oppressed groups - women, LBGT people and people from black and ethnic minorities - to benefit from both progressive legislation and changing social views; it is no coincidence that Mill, eight years after the publication of On liberty, proposed an (unsuccessful) amendment to the Reform Bill that would have extended the franchise to women (Collini 1989: xxviii). Mill, as the arch-proponent of the right to dissent, is committed to the right for people to conduct their 'experiments in living' provided those experiments do not harm others.

Yet some of the beneficiaries of Mill's intellectual inheritance - universities and their unions - now seek to stifle some of the central tenets of On liberty. This is the Paradox of Liberal Inheritance. The City University students' union motion discussed above, and the no-platforming of controversial speakers at universities, is not only contrary to the spirit of On liberty but in outright contradiction. 'If all mankind minus one, were of one opinion, and only one person were of the contrary opinion, mankind would be no more justified in silencing that one person, than he, if he had the power, would be justified in silencing mankind' (Mill 1989: 20). This is a statement by Mill, the arch-liberal. But how many professed liberals would concur with the statement now? Is it, as Bromwich suggests (2016), too liberal a sentiment for contemporary liberals to swallow? This points to the radical nature of On liberty, a radicalness that is sometimes overlooked. Mill is opposed to tyranny in all its forms: the tyranny of the state and the tyranny of the law, but also the tyranny of public opinion. From the following passage, it is more than tempting to think that Mill would view the modern campus as a form of tyranny.

Protection, therefore, against the tyranny of the magistrate is not enough: there needs protection also against the tyranny of the prevailing opinion and feeling; against the tendency of society to impose, by other means than civil penalties, its own ideas and practices as rules of conduct on those who dissent from them; to fetter the development, and, if possible, prevent the formation, of any individuality not in harmony with its ways, and compel all characters to fashion themselves upon the model of its own (1989: 8).

Mill has a tonic - or perhaps, rather, a challenge - for the students' union and the no-platformers: if you cannot defend your position in the teeth of challenge and contradiction from those whose views you disagree with or even detest, then you do not know the grounds for your position. If you cannot stand up and defend your position then your belief is little better than a superstition; an unsupported tenet drifting about in the intellectual ether. It is a prejudice, underpinned by no understanding or rational basis. 'There is the greatest difference between presuming an opinion to be true, because, with every opportunity for contesting it, it has not been refuted, and assuming its truth for the purpose of not permitting its refutation' (1989: 23). Mill, therefore, serves to provide an electric shock for debates about freedom of expression on campus. On liberty also challenges professed liberals to ponder what it truly means to be a liberal.

Why should media students study Mill?

Mill's ideas can challenge thinking on campus in another way. Through his arguments about what constitutes harm, Mill is a powerful lens through which today's media students - tomorrow's custodians of the flow of information - can debate the limits of freedom of expression and regulation, and as such a strong case can be made for his works being a cornerstone for classes on media ethics. On liberty is not the only work of Mill's to consider in this regard. His Utilitarianism (Mill 2011 [1861]) and its refined version of the 'greatest happiness of the greatest number' principle (Gray 2012) is another work that provides an accessible ethical principle from which questions of media ethics can be approached.

There are two other reasons why it is important to use Mill as a means of teaching ethics to students. As a committed reformer and progressive, the starting point of Mill's thinking in many ways chimes with the instinctive intellectual position of many students. Among other causes, Mill argued in favour of votes for women, for the right of public meeting in Hyde Park and for proportional representation (Berlin 2012: 137). Explaining this to students can serve to make Mill seem real and relevant - and therefore worth a listen. Moreover, Mill informs the etiquette and ethics of intellectual debate. Students have been accused of belonging to 'Generation Snowflake' and of 'melting' at the slightest challenge to their opinions (Fox 2016). But Mill presents powerful arguments for why opinions should be held up to ongoing and vigorous scrutiny, and these arguments can be used to emphasise the importance of lively debate in the classroom. Mill believes that it is necessary for even the most secure and sacrosanct beliefs to be held up to scrutiny, otherwise the meaning of such doctrines becomes 'enfeebled', and the risk is that 'living beliefs' become 'dead dogma' (1989: 37). 'Both teachers and learners go to sleep at their post, as soon as there is no enemy in the field' (ibid: 44). Mill is also wary of how the call for 'fair discussion' is open to abuse by being used to shut down vigorous debate. This Millian point has arisen more recently with concerns that pleas for 'civility' in intellectual debate are potentially tantamount to shutting down that debate (Bromwich 2016). Considering the proposition that there should be freedom of opinion on the condition that the delivery should 'not pass the bounds of fair discussion', Mill is cautious.

Much might be said on the importance of fixing where these supposed bounds are to be placed; for if the test be offence to those whose opinion is attacked, I think experience testifies that this offence is given whenever the attack is telling and powerful, and that every opponent who pushes them hard, and whom they find it difficult to answer, appears to them, if he shows any strong feeling on the subject, an intemperate opponent (1989: 54).

As such, Mill has much to say about the tone of healthy debate as much as the content of healthy debate.

Mill and alethic disruption

Earlier, this paper identified and elaborated on two aspects to Mill's thinking in On liberty. Firstly, the notion that stifling free expression is potentially a trans-generational wrong and, secondly, how intolerance - and therefore the restriction of free speech - is the default condition of mankind. The paper also invoked the concept of alethic disruption to describe and account for the spread of 'fake news', affective news and misinformation. Having discussed the application of Mill's arguments to issues of censorship and self-censorship, the paper now ends with a suggestion of how a Millian approach can be applied to the regulation of information in a 'post-truth' era. This argument builds on points made by Jarvis (2016a, 2016b).

As has been discussed, Mill sets a high threshold for what constitutes harm. If an action will cause a physical harm, or in all reasonable likelihood cause a physical harm, to a specific individual or group, then the action can be restricted on that basis. While some malicious items of fake news that equate to incitement might be restricted under this criterion, most would not. To censor such content could be to raise the spectre of a return to the 'default condition of mankind'. Indeed, the presence of falsehood is sometimes welcomed by Mill as a means of jolting truth into life (Mill 1989: 44). But how should the spread of bogus information and demagoguery be countered? For Mill, the answer would be to pro-actively question the falsehoods; to demand that evidence and opinion be given to back them up. Vigorous questioning, after all, is how Mill believes we are able to keep alive 'a living truth' rather than allowing such statements to ossify into 'dead dogma' (ibid: 37).

A Millian approach would, therefore, be to take on the purveyors of misinformation in their own arena; to take to the social media platforms with facts. As Jarvis puts it, the answer is to 'flood the zone with good information' (2016a) rather than seeking to censor what is misinformation. But to do this, journalists can no longer just produce content in traditional column inches or long-form articles on web pages, otherwise they - and their fact-checking endeavours - face irrelevance. Instead, traditional media 'must adapt to their new reality and bring their journalism - their facts, fact-checking, reporting, explanation and context - to the public where the public is, in a form and voice that is appropriate to the context and use of each platform' (Jarvis 2016b). This means utilising memes, GIFs, video and other online techniques that engage social media users. Jarvis contends that it is impossible to play 'Whac-A-Mole' with inaccurate information, by which he means scouring the web and trying to delete each and every piece of bogus content when it pops up.

Instead of mourning the creation of fake-news memes and putting the onus on Facebook to kill them… we should be pouring out our own truth memes - with facts, fact-checking, context, explanation, education, reporting, watch-dogging: journalism, in short. We should be arming fair-minded, intelligent, curious, rational, fact-loving citizens … with the weapons, the truth bullets, to fire at will in their conversations. They won't win all the wars but they will win some fact battles alongside us if only we enable them (Jarvis 2016b).

Although Jarvis does not mention Mill in his argument it is, nonetheless, Millian with its emphasis on truth doing battle in an unrestricted market place of ideas. As such, it is another illustration of how Mill's legacy continues to indirectly inform contemporary issues of free expression.

Conclusion

Despite writing in Victorian times, Mill outlines positions and arguments that modern academe can usefully ponder amid concerns over on-campus censorship, and self-censorship more widely. This paper has suggested that, while contentious in its application in particular instances, the Harm Principle remains a powerful starting point for contemporary debates about free expression. Moreover, it has been argued that Mill articulates principles that modern media students can use as an accessible means of approaching questions of media ethics. He also makes powerful points about the etiquette of intellectual debate. In addition, it has been argued that Mill's vision for a liberal society and the implications of that liberal creed on free expression pose challenging questions for those who currently profess to be liberals. Despite his arguments having been made more than a century-and-a-half ago, the paper has explored how Mill speaks engagingly to current debates over freedom of expression.

Note on the contributor

Tom Bradshaw is Senior Lecturer at the University of Gloucestershire's School of Media, where he is course leader for the BA (Hons.) Sports Journalism programme. He studied philosophy at the University of Cambridge and then journalism at the University of Central Lancashire before becoming a news and sports journalist. He has worked as a print, online and radio journalist for organisations across the UK and South Africa. He currently works as a freelance journalist and broadcaster for the BBC and The Times, among others, and has contributed to The Times Literary Supplement. His research interests include uses of technology in education and media ethics, particularly self-censorship.